From 50d68427b7659d709e0ae686cd5acfbca64ec592 Mon Sep 17 00:00:00 2001 From: Jeff Otterson Date: Thu, 26 Feb 2026 09:13:17 -0700 Subject: [PATCH] Add recruiting plugin for talent acquisition workflows Co-Authored-By: Claude Opus 4.6 --- .claude-plugin/marketplace.json | 5 + README.md | 3 +- recruiting/.claude-plugin/plugin.json | 9 ++ recruiting/.mcp.json | 52 +++++++ recruiting/CONNECTORS.md | 22 +++ recruiting/README.md | 143 ++++++++++++++++++ recruiting/commands/debrief.md | 82 ++++++++++ recruiting/commands/interview-prep.md | 94 ++++++++++++ recruiting/commands/offer-prep.md | 101 +++++++++++++ recruiting/commands/outreach.md | 86 +++++++++++ recruiting/commands/pipeline-review.md | 81 ++++++++++ recruiting/commands/screen.md | 75 +++++++++ recruiting/commands/source.md | 82 ++++++++++ recruiting/commands/write-jd.md | 83 ++++++++++ .../skills/candidate-evaluation/SKILL.md | 76 ++++++++++ .../references/red-flags-green-flags.md | 60 ++++++++ .../references/scoring-rubrics.md | 72 +++++++++ .../skills/diversity-recruiting/SKILL.md | 105 +++++++++++++ .../references/inclusive-language-guide.md | 107 +++++++++++++ recruiting/skills/employer-branding/SKILL.md | 101 +++++++++++++ recruiting/skills/interview-design/SKILL.md | 84 ++++++++++ .../references/interview-formats.md | 107 +++++++++++++ .../references/question-bank.md | 110 ++++++++++++++ recruiting/skills/job-architecture/SKILL.md | 71 +++++++++ .../references/leveling-frameworks.md | 78 ++++++++++ .../skills/market-intelligence/SKILL.md | 86 +++++++++++ recruiting/skills/pipeline-analytics/SKILL.md | 90 +++++++++++ recruiting/skills/sourcing-strategy/SKILL.md | 113 ++++++++++++++ 28 files changed, 2177 insertions(+), 1 deletion(-) create mode 100644 recruiting/.claude-plugin/plugin.json create mode 100644 recruiting/.mcp.json create mode 100644 recruiting/CONNECTORS.md create mode 100644 recruiting/README.md create mode 100644 recruiting/commands/debrief.md create mode 100644 recruiting/commands/interview-prep.md create mode 100644 recruiting/commands/offer-prep.md create mode 100644 recruiting/commands/outreach.md create mode 100644 recruiting/commands/pipeline-review.md create mode 100644 recruiting/commands/screen.md create mode 100644 recruiting/commands/source.md create mode 100644 recruiting/commands/write-jd.md create mode 100644 recruiting/skills/candidate-evaluation/SKILL.md create mode 100644 recruiting/skills/candidate-evaluation/references/red-flags-green-flags.md create mode 100644 recruiting/skills/candidate-evaluation/references/scoring-rubrics.md create mode 100644 recruiting/skills/diversity-recruiting/SKILL.md create mode 100644 recruiting/skills/diversity-recruiting/references/inclusive-language-guide.md create mode 100644 recruiting/skills/employer-branding/SKILL.md create mode 100644 recruiting/skills/interview-design/SKILL.md create mode 100644 recruiting/skills/interview-design/references/interview-formats.md create mode 100644 recruiting/skills/interview-design/references/question-bank.md create mode 100644 recruiting/skills/job-architecture/SKILL.md create mode 100644 recruiting/skills/job-architecture/references/leveling-frameworks.md create mode 100644 recruiting/skills/market-intelligence/SKILL.md create mode 100644 recruiting/skills/pipeline-analytics/SKILL.md create mode 100644 recruiting/skills/sourcing-strategy/SKILL.md diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.json index a82ab1c..7d303cc 100644 --- a/.claude-plugin/marketplace.json +++ b/.claude-plugin/marketplace.json @@ -49,6 +49,11 @@ "source": "./customer-support", "description": "Triage tickets, draft responses, escalate issues, and build your knowledge base. Research customer context and turn resolved issues into self-service content." }, + { + "name": "recruiting", + "source": "./recruiting", + "description": "Screen resumes, write job descriptions, source candidates, build interview kits, run debriefs, and manage your hiring pipeline. Works standalone or supercharged with ATS, email, and calendar integrations." + }, { "name": "product-management", "source": "./product-management", diff --git a/README.md b/README.md index c902fcb..584318c 100644 --- a/README.md +++ b/README.md @@ -10,13 +10,14 @@ Each plugin bundles the skills, connectors, slash commands, and sub-agents for a ## Plugin Marketplace -We're open-sourcing 11 plugins built and inspired by our own work: +We're open-sourcing 12 plugins built and inspired by our own work: | Plugin | How it helps | Connectors | |--------|-------------|------------| | **[productivity](./productivity)** | Manage tasks, calendars, daily workflows, and personal context so you spend less time repeating yourself. | Slack, Notion, Asana, Linear, Jira, Monday, ClickUp, Microsoft 365 | | **[sales](./sales)** | Research prospects, prep for calls, review your pipeline, draft outreach, and build competitive battlecards. | Slack, HubSpot, Close, Clay, ZoomInfo, Notion, Jira, Fireflies, Microsoft 365 | | **[customer-support](./customer-support)** | Triage tickets, draft responses, package escalations, research customer context, and turn resolved issues into knowledge base articles. | Slack, Intercom, HubSpot, Guru, Jira, Notion, Microsoft 365 | +| **[recruiting](./recruiting)** | Screen resumes, source candidates, design interviews, run debriefs, write job descriptions, and manage your hiring pipeline. | Slack, Greenhouse, Lever, Apollo, Clay, LinkedIn, Notion, Zoom, Google Calendar, Gmail | | **[product-management](./product-management)** | Write specs, plan roadmaps, synthesize user research, keep stakeholders updated, and track the competitive landscape. | Slack, Linear, Asana, Monday, ClickUp, Jira, Notion, Figma, Amplitude, Pendo, Intercom, Fireflies | | **[marketing](./marketing)** | Draft content, plan campaigns, enforce brand voice, brief on competitors, and report on performance across channels. | Slack, Canva, Figma, HubSpot, Amplitude, Notion, Ahrefs, SimilarWeb, Klaviyo | | **[legal](./legal)** | Review contracts, triage NDAs, navigate compliance, assess risk, prep for meetings, and draft templated responses. | Slack, Box, Egnyte, Jira, Microsoft 365 | diff --git a/recruiting/.claude-plugin/plugin.json b/recruiting/.claude-plugin/plugin.json new file mode 100644 index 0000000..2314272 --- /dev/null +++ b/recruiting/.claude-plugin/plugin.json @@ -0,0 +1,9 @@ +{ + "name": "recruiting", + "version": "1.0.0", + "description": "Screen resumes, write job descriptions, source candidates, build interview kits, run debriefs, and manage your hiring pipeline. Built for talent acquisition professionals, HR generalists, and startup founders making their first hires.", + "author": { + "name": "Jeff Otterson", + "url": "https://meritforgeai.com" + } +} diff --git a/recruiting/.mcp.json b/recruiting/.mcp.json new file mode 100644 index 0000000..75f8091 --- /dev/null +++ b/recruiting/.mcp.json @@ -0,0 +1,52 @@ +{ + "mcpServers": { + "slack": { + "type": "http", + "url": "https://mcp.slack.com/mcp" + }, + "google-calendar": { + "type": "http", + "url": "https://gcal.mcp.claude.com/mcp" + }, + "gmail": { + "type": "http", + "url": "https://gmail.mcp.claude.com/mcp" + }, + "ms365": { + "type": "http", + "url": "https://microsoft365.mcp.claude.com/mcp" + }, + "notion": { + "type": "http", + "url": "https://mcp.notion.com/mcp" + }, + "atlassian": { + "type": "http", + "url": "https://mcp.atlassian.com/v1/mcp" + }, + "greenhouse": { + "type": "http", + "url": "https://mcp.greenhouse.io/mcp" + }, + "lever": { + "type": "http", + "url": "https://mcp.lever.co/mcp" + }, + "apollo": { + "type": "http", + "url": "https://api.apollo.io/mcp" + }, + "linkedin": { + "type": "http", + "url": "https://mcp.linkedin.com/mcp" + }, + "clay": { + "type": "http", + "url": "https://api.clay.com/v3/mcp" + }, + "zoom": { + "type": "http", + "url": "https://mcp.zoom.us/mcp" + } + } +} diff --git a/recruiting/CONNECTORS.md b/recruiting/CONNECTORS.md new file mode 100644 index 0000000..06627dc --- /dev/null +++ b/recruiting/CONNECTORS.md @@ -0,0 +1,22 @@ +# Connectors + +## How tool references work + +Plugin files use `~~category` as a placeholder for whatever tool the user connects in that category. For example, `~~ATS` might mean Greenhouse, Lever, Ashby, or any other applicant tracking system with an MCP server. + +Plugins are **tool-agnostic** — they describe workflows in terms of categories (ATS, email, calendar, etc.) rather than specific products. The `.mcp.json` pre-configures specific MCP servers, but any MCP server in that category works. + +## Connectors for this plugin + +| Category | Placeholder | Included servers | Other options | +|----------|-------------|-----------------|---------------| +| ATS | `~~ATS` | Greenhouse, Lever | Ashby, Workday Recruiting, iCIMS, Jobvite, SmartRecruiters | +| Calendar | `~~calendar` | Google Calendar, Microsoft 365 | Calendly | +| Chat | `~~chat` | Slack | Microsoft Teams, Discord | +| Cloud storage | `~~cloud storage` | — | Google Drive, Dropbox, SharePoint | +| CRM | `~~CRM` | — | HubSpot, Salesforce, Bullhorn | +| Data enrichment | `~~data enrichment` | Apollo, Clay, LinkedIn | Clearbit, ZoomInfo, Lusha, People Data Labs | +| Email | `~~email` | Gmail, Microsoft 365 | — | +| HRIS | `~~HRIS` | — | BambooHR, Workday, Rippling, Gusto | +| Knowledge base | `~~knowledge base` | Notion, Atlassian (Confluence) | Guru, Slite, Coda | +| Video conferencing | `~~video` | Zoom | Google Meet, Microsoft Teams | diff --git a/recruiting/README.md b/recruiting/README.md new file mode 100644 index 0000000..5665b1e --- /dev/null +++ b/recruiting/README.md @@ -0,0 +1,143 @@ +# Recruiting Plugin + +A recruiting and talent acquisition plugin primarily designed for [Cowork](https://claude.com/product/cowork), Anthropic's agentic desktop application — though it also works in Claude Code. Source candidates, screen resumes, design interviews, run debriefs, write job descriptions, and manage your hiring pipeline. Works with any recruiting team — standalone with web search and your input, supercharged when you connect your ATS, email, and other tools. + +## Installation + +```bash +claude plugins add knowledge-work-plugins/recruiting +``` + +## Commands + +Explicit workflows you invoke with a slash command: + +| Command | Description | +|---|---| +| `/source` | Build a targeted sourcing strategy — channels, Boolean strings, outreach angles | +| `/screen` | Screen a candidate against job requirements — fit matrix, signals, and recommendation | +| `/interview-prep` | Generate interview questions, scorecards, and loop structure for a role | +| `/debrief` | Synthesize interview feedback into a structured hiring recommendation | +| `/pipeline-review` | Analyze pipeline health — conversion rates, bottlenecks, and action plan | +| `/write-jd` | Write or improve a job description with inclusive language and clear requirements | +| `/outreach` | Draft personalized candidate outreach messages with research-backed hooks | +| `/offer-prep` | Prepare and analyze a compensation offer with market benchmarks | + +All commands work **standalone** (paste resumes, JDs, notes, or describe your situation) and get **supercharged** with MCP connectors. + +## Skills + +Domain knowledge Claude uses automatically when relevant: + +| Skill | Description | +|---|---| +| `candidate-evaluation` | Structured rubrics and scorecards for consistent, evidence-based assessment with bias mitigation | +| `interview-design` | Build interview loops with competency coverage mapping and format selection | +| `job-architecture` | Leveling frameworks, competency models, and role scoping across functions | +| `market-intelligence` | Talent market conditions, compensation benchmarks, and competitor intel | +| `pipeline-analytics` | Funnel metrics, conversion rates, time-to-fill analysis, and bottleneck diagnosis | +| `sourcing-strategy` | Boolean strings, channel mix optimization, and passive candidate engagement | +| `diversity-recruiting` | Inclusive hiring practices, bias mitigation, and diverse sourcing strategies | +| `employer-branding` | EVP messaging, job marketing, and candidate experience optimization | + +## Example Workflows + +### Writing a Job Description + +``` +/write-jd +``` + +Describe the role you're hiring for. Get a polished job description with clear requirements, inclusive language, and a compelling pitch. If your knowledge base is connected, it pulls your company's JD template and values. + +### Screening a Candidate + +``` +/screen +``` + +Paste a resume and job description. Get a structured fit matrix scoring every dimension, strong signals, gaps, recruiter screen questions, and a clear advance/hold/pass recommendation. + +### Building a Sourcing Plan + +``` +/source +``` + +Describe the role and target profile. Get a multi-channel sourcing strategy with Boolean search strings, outreach templates, and channel prioritization based on the talent market. + +### Running a Debrief + +``` +/debrief +``` + +Paste interview feedback from your panel. Get a synthesized recommendation with dimension-by-dimension analysis, consensus view, and a clear hire/no-hire recommendation with reasoning. + +### Researching the Market + +Just ask naturally: +``` +What's the market rate for a senior backend engineer in Austin? +``` + +The `market-intelligence` skill triggers automatically and gives you compensation benchmarks, talent supply data, and competitive landscape. + +### Improving Your Pipeline + +``` +/pipeline-review +``` + +Paste pipeline data or describe your funnel. Get conversion rates benchmarked against industry standards, bottleneck diagnosis, and a prioritized action plan. + +## Standalone + Supercharged + +Every command and skill works without any integrations: + +| What You Can Do | Standalone | Supercharged With | +|-----------------|------------|-------------------| +| Screen candidates | Paste resume + JD | ATS MCP (e.g. Greenhouse, Lever) | +| Write job descriptions | Describe the role | Knowledge base MCP (templates, values) | +| Build sourcing plans | Describe target profile | Enrichment MCP (e.g. Apollo, Clay) | +| Design interviews | Describe role + competencies | Knowledge base MCP (rubrics, guides) | +| Run debriefs | Paste interview feedback | ATS MCP (scorecards, history) | +| Draft outreach | Web search + your context | Email, Enrichment MCPs | +| Analyze pipeline | Paste data or describe funnel | ATS MCP (live pipeline data) | +| Prep offers | Describe role + candidate | HRIS MCP (comp bands, approvals) | + +## MCP Integrations + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](CONNECTORS.md). + +Connect your tools for a richer experience: + +| Category | Examples | What It Enables | +|---|---|---| +| **ATS** | Greenhouse, Lever | Live candidate pipelines, stage updates, scorecards, duplicate detection | +| **Enrichment** | Apollo, Clay, LinkedIn | Candidate profile enrichment, contact info, company research | +| **Chat** | Slack, Teams | Hiring channel updates, hiring manager notifications, scorecard sharing | +| **Calendar** | Google Calendar, Microsoft 365 | Interviewer availability, loop scheduling, invite management | +| **Email** | Gmail, Microsoft 365 | Candidate outreach, follow-ups, offer letters | + +See [CONNECTORS.md](CONNECTORS.md) for the full list of supported integrations, including HRIS, knowledge base, and video conferencing options. + +## Settings + +Create a local settings file at `recruiting/.claude/settings.local.json` to personalize: + +```json +{ + "company": "Your Company", + "industry": "Your Industry", + "team_size": "Approximate headcount", + "hiring_bar": "Key attributes your company values", + "interview_process": "Your standard interview loop structure", + "tools": { + "ats": "Greenhouse", + "hris": "Workday" + } +} +``` + +The plugin will ask you for this information interactively if it's not configured. diff --git a/recruiting/commands/debrief.md b/recruiting/commands/debrief.md new file mode 100644 index 0000000..5c758aa --- /dev/null +++ b/recruiting/commands/debrief.md @@ -0,0 +1,82 @@ +--- +description: Synthesize interview feedback into a hiring recommendation +argument-hint: "" +--- + +# Interview Debrief + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Synthesize feedback from multiple interviewers into a structured hiring recommendation. Surfaces consensus, disagreements, and evidence gaps to drive a productive debrief conversation. + +## Input + +The user provides one or more of: +- Individual interviewer scorecards or notes (pasted, uploaded, or described) +- Candidate name and role +- Interview panel composition +- Any known concerns or highlights from the hiring manager + +If no feedback is provided, ask: "Share the interview feedback — paste scorecards, notes from each interviewer, or describe the signals from the panel." + +## Workflow + +1. **Collect feedback** — Parse all interviewer input, attributing each signal to its source +2. **Map to competencies** — Organize feedback by the competencies being evaluated, not by interviewer +3. **Identify consensus** — Flag areas where all interviewers agree (both positive and negative) +4. **Surface disagreements** — Highlight competencies where interviewers diverge, with each side's evidence +5. **Find evidence gaps** — Note competencies that were insufficiently tested or have conflicting signals +6. **Assess against the bar** — Compare the composite signal against the role's hiring bar +7. **Generate recommendation** — Produce a clear recommendation with confidence level +8. **Prepare debrief agenda** — Structure the debrief discussion to focus on disagreements and gaps first + +## Output Structure + +``` +## Debrief Summary: [Candidate Name] → [Role Title] + +### Panel +| Interviewer | Stage | Overall rating | +|-------------|-------|---------------| +| [Name] | [Stage] | [Rating] | +| ... | ... | ... | + +### Competency Heatmap + +| Competency | [Interviewer 1] | [Interviewer 2] | [Interviewer 3] | Consensus | +|------------|----------------|----------------|----------------|-----------| +| [Comp 1] | ✅ Strong | ✅ Positive | ⚠️ Mixed | Positive | +| [Comp 2] | ❌ Concern | ⚠️ Mixed | ❌ Concern | Concern | +| ... | ... | ... | ... | ... | + +### Areas of Consensus +**Strengths**: [What everyone agrees is strong, with evidence] +**Concerns**: [What everyone agrees is a gap, with evidence] + +### Areas of Disagreement +**[Competency]**: [Interviewer A] saw [X signal] while [Interviewer B] saw [Y signal]. This likely reflects [possible explanation]. **Discuss in debrief.** + +### Evidence Gaps +- [Competency not adequately tested — suggest follow-up] + +### Recommendation: [Strong Hire / Hire / No Hire / Strong No Hire] +**Confidence**: [High / Medium / Low] +**Reasoning**: [2-3 sentences synthesizing the overall signal] + +### Debrief Discussion Guide +1. **Start with**: [Key disagreement to resolve] +2. **Then discuss**: [Evidence gap to address] +3. **Decision point**: [What would move this from X to Y?] +``` + +## With Connectors + +- **If ATS connected**: Pull submitted scorecards directly, update candidate disposition after the debrief +- **If chat connected**: Post the debrief summary to the hiring channel, notify the hiring manager +- **If calendar connected**: Check the scheduled debrief time and attendees + +## Tips + +- Best results when you have written feedback from every interviewer in the loop +- Include the role's leveling expectations if the decision hinges on "is this a senior or staff?" +- Flag if any interviewer has a known bias toward certain backgrounds or styles diff --git a/recruiting/commands/interview-prep.md b/recruiting/commands/interview-prep.md new file mode 100644 index 0000000..f1df2a3 --- /dev/null +++ b/recruiting/commands/interview-prep.md @@ -0,0 +1,94 @@ +--- +description: Generate interview questions and scorecards +argument-hint: "" +--- + +# Interview Prep + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Generate a complete interview kit: structured questions, evaluation rubrics, and scorecards tailored to a specific role and interview stage. + +## Input + +The user provides one or more of: +- Role title and level +- Interview stage (phone screen, technical, behavioral, hiring manager, executive, culture) +- Specific competencies to assess +- Job description +- Candidate resume (to customize questions) +- Interview duration + +If minimal input, ask: "What role and interview stage are you prepping for? (e.g., 'Senior Backend Engineer - technical interview, 60 minutes')" + +## Workflow + +1. **Define the interview scope** — Identify the stage, duration, competencies to assess, and interviewer role +2. **Select competencies** — Map 3-5 competencies this interview should evaluate, avoiding overlap with other stages if the user describes the full loop +3. **Generate questions** — Write 4-6 primary questions with follow-up probes, calibrated to the role level +4. **Build scoring rubric** — Define what "strong hire", "hire", "no hire", and "strong no hire" look like for each competency +5. **Create the scorecard** — Structured template the interviewer can fill out during or after the interview +6. **Add interviewer guidance** — Time allocation, red/green flags to watch for, and common biases to avoid +7. **If candidate resume provided** — Customize questions to probe specific claims, career transitions, or gaps + +## Output Structure + +``` +## Interview Kit: [Role] — [Stage] +**Duration**: [X minutes] | **Competencies**: [list] + +### Opening (5 min) +- Introduce yourself and the role +- Set expectations for the interview format + +### Questions + +#### Q1: [Question text] ([competency], [X min]) +**Follow-ups**: +- [Probe deeper on X] +- [Ask for a counter-example] + +| Rating | Signal | +|--------|--------| +| Strong hire | [specific observable behavior] | +| Hire | [specific observable behavior] | +| No hire | [specific observable behavior] | +| Strong no hire | [specific observable behavior] | + +#### Q2: [Question text] ([competency], [X min]) +... + +### Scorecard + +| Competency | Rating (1-4) | Notes | +|------------|-------------|-------| +| [Competency 1] | ___ | | +| [Competency 2] | ___ | | +| ... | | | + +**Overall recommendation**: Strong hire / Hire / No hire / Strong no hire + +**Key evidence for recommendation**: + +### Interviewer Guidance +- **Time management**: [allocation suggestions] +- **Green flags**: [what to look for] +- **Red flags**: [concerning signals] +- **Bias watch**: [common biases for this interview type] + +### Closing (5 min) +- Allow candidate questions +- Explain next steps and timeline +``` + +## With Connectors + +- **If ATS connected**: Pull the interview plan, see what other interviewers are covering, and submit the scorecard directly +- **If knowledge base connected**: Reference the company's competency framework, leveling rubrics, and past interview guides +- **If calendar connected**: Check the scheduled interview time and interviewer details + +## Tips + +- Specify the full interview loop if possible so questions don't overlap across stages +- Include the candidate's resume for targeted questions that validate specific claims +- For panel interviews, note how many interviewers and their roles diff --git a/recruiting/commands/offer-prep.md b/recruiting/commands/offer-prep.md new file mode 100644 index 0000000..7e588c3 --- /dev/null +++ b/recruiting/commands/offer-prep.md @@ -0,0 +1,101 @@ +--- +description: Prepare and analyze a compensation offer +argument-hint: "" +--- + +# Offer Prep + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Build a competitive compensation offer with market analysis, internal equity considerations, and negotiation guidance. Produces a complete offer package recommendation. + +## Input + +The user provides one or more of: +- Role title and level +- Candidate name and current compensation (if known) +- Candidate's expectations or competing offers +- Company compensation bands or philosophy +- Location (for geo-adjusted comp) +- Equity/stock component details +- Sign-on bonus budget + +If minimal input, ask: "What role and level is the offer for? Share any comp data you have — bands, candidate expectations, or competing offers." + +## Workflow + +1. **Define the offer parameters** — Role, level, location, and any constraints from `$ARGUMENTS` +2. **Research market rates** — Use web search for comp benchmarks at this role/level/location (levels.fyi, Glassdoor, Blind, Pave, Radford data) +3. **Analyze the candidate's position** — Current comp, competing offers, and likely expectations +4. **Build the package** — Base salary, equity, bonus, sign-on, and benefits mapped against market data +5. **Assess internal equity** — Flag if this offer would create compression or inversion risks with existing team members +6. **Model scenarios** — Present a low/mid/high range with trade-offs for each +7. **Prepare negotiation guidance** — Anticipate likely counterpoints and prepare responses +8. **If HRIS connected** — Pull approved compensation bands, existing team comp data, and benefits details + +## Output Structure + +``` +## Offer Analysis: [Candidate Name] → [Role Title, Level] +**Location**: [Location] | **Market**: [market characterization] + +### Market Benchmarks + +| Component | 25th %ile | 50th %ile | 75th %ile | 90th %ile | +|-----------|----------|----------|----------|----------| +| Base salary | $[X] | $[X] | $[X] | $[X] | +| Total cash (base + bonus) | $[X] | $[X] | $[X] | $[X] | +| Total comp (incl. equity) | $[X] | $[X] | $[X] | $[X] | + +*Sources: [list data sources used]* + +### Recommended Offer + +| Component | Recommended | Range | Notes | +|-----------|------------|-------|-------| +| Base salary | $[X] | $[low]-$[high] | [reasoning] | +| Annual bonus | $[X] ([%]) | — | [target %] | +| Equity | [X shares/units] | — | [vesting schedule, current value] | +| Sign-on bonus | $[X] | — | [if applicable, why] | +| **Total Year 1** | **$[X]** | | | +| **Annual ongoing** | **$[X]** | | | + +### Candidate Analysis +- **Current comp**: [if known] +- **Expected uplift**: [typical move premium is 10-20%] +- **Competing offers**: [if known] +- **Offer positioning**: [where this sits relative to their expectations] + +### Internal Equity Check +- **Team range for this level**: [if known] +- **Compression risk**: [Yes/No — details] +- **Recommendation**: [any adjustments needed] + +### Negotiation Playbook +| If they say... | Respond with... | Lever to pull | +|---------------|----------------|---------------| +| "Base is too low" | [response] | [equity, sign-on, review cycle] | +| "I have a competing offer at $X" | [response] | [total comp comparison, non-monetary value] | +| "I need more equity" | [response] | [refresher schedule, grant structure] | + +### Approval Checklist +- [ ] Within approved compensation band +- [ ] Internal equity reviewed +- [ ] Hiring manager sign-off +- [ ] Total comp budget approved +- [ ] Start date confirmed +``` + +## With Connectors + +- **If HRIS connected**: Pull compensation bands, team comp data, and benefits package details +- **If ATS connected**: Update the candidate to "offer" stage, attach the offer analysis +- **If email connected**: Draft the offer letter email +- **If knowledge base connected**: Pull the offer letter template, benefits one-pager, and equity plan details + +## Tips + +- Always present total compensation, not just base salary +- Include the candidate's current comp if available — it anchors the negotiation analysis +- Mention competing offers or other leverage the candidate has for realistic guidance +- For equity-heavy offers, show the value at different stock price scenarios diff --git a/recruiting/commands/outreach.md b/recruiting/commands/outreach.md new file mode 100644 index 0000000..55015eb --- /dev/null +++ b/recruiting/commands/outreach.md @@ -0,0 +1,86 @@ +--- +description: Draft personalized candidate outreach messages +argument-hint: "" +--- + +# Draft Outreach + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Craft personalized outreach messages that stand out in a candidate's inbox. Generates multi-touch sequences with tailored hooks based on the candidate's background and motivations. + +## Input + +The user provides one or more of: +- Candidate name, current role, and company +- LinkedIn profile or resume +- Role being recruited for +- Why this candidate is a fit (specific reasons) +- Outreach channel (email, LinkedIn InMail, other) +- Any prior relationship or warm connection + +If minimal input, ask: "Who are you reaching out to and for what role? Share the candidate's background and the position." + +## Workflow + +1. **Research the candidate** — Parse provided info and use web search to understand their background, interests, recent activity, and likely motivations +2. **Identify hooks** — Find 2-3 personalized angles: recent project, company news, shared connection, career trajectory, technical interests +3. **Match to role** — Connect the candidate's likely motivations to what the role offers +4. **Draft the sequence** — Write a 3-touch outreach sequence (initial + 2 follow-ups) with different angles +5. **Optimize for channel** — Adjust length and tone for the outreach channel (LinkedIn InMail is shorter, email allows more detail) +6. **Add subject lines** — Write 2-3 subject line options for email outreach +7. **If data enrichment connected** — Pull verified contact info, mutual connections, and recent activity + +## Output Structure + +``` +## Outreach: [Candidate Name] → [Role Title] + +### Candidate Research +- **Current**: [Role at Company] +- **Background**: [Key career highlights] +- **Likely motivations**: [What would make them move] +- **Personalization hooks**: [Specific details to reference] + +### Message 1: Initial Outreach +**Channel**: [Email/LinkedIn] | **Tone**: [warm/direct/curious] +**Subject line options**: +1. [Option A] +2. [Option B] + +--- +[Message body — personalized, concise, specific about why them + why this role] +--- + +### Message 2: Follow-up (3-5 days later) +**Angle**: [Different hook than message 1] + +--- +[Follow-up message — adds new information, doesn't just "bump"] +--- + +### Message 3: Final Touch (5-7 days later) +**Angle**: [Lightest touch, easy call-to-action] + +--- +[Brief final message — respectful, offers a clear next step or graceful close] +--- + +### Outreach Tips +- **Best time to send**: [day/time recommendation] +- **If they respond positively**: [suggested next steps] +- **If no response after sequence**: [recommended waiting period and re-engagement strategy] +``` + +## With Connectors + +- **If email connected**: Send the messages directly, schedule follow-ups +- **If data enrichment connected**: Pull verified email, phone, mutual connections, and recent social activity +- **If ATS connected**: Log the outreach, check for prior contact history, and avoid double-outreach +- **If chat connected**: Notify the hiring manager when a candidate responds + +## Tips + +- The more you know about the candidate, the better the personalization — LinkedIn profiles produce much better outreach than just a name +- Mention the specific reason you think they're a fit — generic "your background is impressive" messages get ignored +- For passive candidates, lead with what's in it for them, not what you need diff --git a/recruiting/commands/pipeline-review.md b/recruiting/commands/pipeline-review.md new file mode 100644 index 0000000..4ecc836 --- /dev/null +++ b/recruiting/commands/pipeline-review.md @@ -0,0 +1,81 @@ +--- +description: Review pipeline health and identify bottlenecks +argument-hint: "" +--- + +# Pipeline Review + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Analyze your recruiting pipeline to surface bottlenecks, stalled candidates, and conversion rate issues. Produces actionable recommendations to improve pipeline velocity. + +## Input + +The user provides one or more of: +- Role or team to review +- Pipeline data (pasted from ATS, spreadsheet, or described) +- Current candidate counts by stage +- Time-in-stage data +- Hiring goals and deadlines + +If no data is provided, ask: "Share your pipeline data — paste candidate counts by stage, export from your ATS, or describe where things stand for a role or team." + +## Workflow + +1. **Parse pipeline data** — Extract candidates by stage, time-in-stage, and any conversion metrics +2. **Calculate funnel metrics** — Conversion rates between stages, pipeline velocity, projected time-to-fill +3. **Identify bottlenecks** — Flag stages with below-benchmark conversion, stalled candidates, or capacity constraints +4. **Assess pipeline health** — Compare current pipeline against what's needed to make the hire on time +5. **Model scenarios** — If conversion rates hold, how many candidates are needed at top-of-funnel? +6. **Generate recommendations** — Prioritized actions to unblock the pipeline +7. **If ATS connected** — Pull live pipeline data, identify specific stalled candidates, and check interviewer capacity + +## Output Structure + +``` +## Pipeline Review: [Role/Team] +**As of**: [date] | **Target hire date**: [date] | **Days open**: [N] + +### Funnel Summary + +| Stage | Candidates | Conversion | Benchmark | Status | +|-------|-----------|-----------|-----------|--------| +| Applied/Sourced | [N] | — | — | | +| Recruiter screen | [N] | [%] | 25-30% | ✅/⚠️/❌ | +| Hiring manager screen | [N] | [%] | 40-50% | ✅/⚠️/❌ | +| Technical/Skills | [N] | [%] | 30-40% | ✅/⚠️/❌ | +| Onsite/Final | [N] | [%] | 40-50% | ✅/⚠️/❌ | +| Offer | [N] | [%] | 70-85% | ✅/⚠️/❌ | +| Hire | [N] | — | — | | + +### Bottlenecks +1. **[Stage]**: [Description of the problem, e.g., "Conversion from screen to HM review is 15% vs 40% benchmark — likely a calibration issue between sourcing criteria and HM expectations"] +2. ... + +### Stalled Candidates +| Candidate | Stage | Days in stage | Recommended action | +|-----------|-------|--------------|-------------------| +| [Name] | [Stage] | [N] | [action] | + +### Pipeline Math +- To make **1 hire** by [date] at current conversion rates, you need **[N] candidates** at top of funnel +- Current pace: [X candidates/week entering] → projected hire by [date] +- Gap: [N more candidates needed] or [improve conversion at stage X] + +### Recommendations +1. **[Priority 1]**: [Specific action with expected impact] +2. **[Priority 2]**: [Specific action] +3. **[Priority 3]**: [Specific action] +``` + +## With Connectors + +- **If ATS connected**: Pull real-time pipeline data, identify stalled candidates by name, check interviewer load +- **If calendar connected**: Assess interviewer availability and scheduling bottlenecks +- **If chat connected**: Post the pipeline summary to the hiring channel + +## Tips + +- Run weekly for active roles to catch problems early +- Include time-in-stage data for the most actionable analysis +- Compare across similar roles to identify systemic vs. role-specific issues diff --git a/recruiting/commands/screen.md b/recruiting/commands/screen.md new file mode 100644 index 0000000..32e99e1 --- /dev/null +++ b/recruiting/commands/screen.md @@ -0,0 +1,75 @@ +--- +description: Screen a candidate against job requirements +argument-hint: "" +--- + +# Screen Candidate + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Evaluate a candidate's fit against specific role requirements. Produces a structured assessment with clear signal on whether to advance, with reasoning for every dimension. + +## Input + +The user provides one or more of: +- Resume (pasted text, uploaded file, or link) +- LinkedIn profile summary +- Candidate notes from a recruiter screen +- Job description or role requirements to screen against + +If no job requirements are provided, ask: "What role are you screening for? Paste the JD or describe the key requirements." + +If no candidate info is provided, ask: "Share the candidate's resume, LinkedIn summary, or any background info you have." + +## Workflow + +1. **Extract requirements** — Parse the job description into must-have qualifications, nice-to-haves, and disqualifiers +2. **Parse candidate profile** — Extract work history, skills, education, achievements, and career trajectory from the provided materials +3. **Assess fit by dimension** — Score each requirement dimension: technical skills, domain experience, seniority level, culture indicators, career trajectory +4. **Identify signals and gaps** — Flag strong positive signals, concerning gaps, and areas needing clarification +5. **Generate recruiter screen questions** — Write 3-5 targeted questions to validate gaps or ambiguous areas +6. **Produce recommendation** — Strong advance / Advance / Hold for discussion / Pass, with clear reasoning +7. **If ATS connected** — Check for prior applications, referral source, and any existing notes + +## Output Structure + +``` +## Candidate Screen: [Candidate Name] → [Role Title] + +### Overall Assessment: [Strong Advance / Advance / Hold / Pass] + +### Fit Matrix + +| Dimension | Requirement | Candidate signal | Fit | +|-----------|------------|-----------------|-----| +| [Technical skill] | [what's needed] | [what they have] | ✅/⚠️/❌ | +| [Domain experience] | ... | ... | ... | +| [Seniority/scope] | ... | ... | ... | +| [Education/certs] | ... | ... | ... | + +### Strong Signals +- [Positive indicator with evidence] + +### Gaps & Risks +- [Gap or concern with context on severity] + +### Recruiter Screen Questions +1. [Question targeting a specific gap] +2. [Question to validate a claim] +3. [Question about motivation/fit] + +### Recommendation +[2-3 sentence summary of recommendation with key reasoning] +``` + +## With Connectors + +- **If ATS connected**: Check application history, source attribution, and prior interview feedback +- **If data enrichment connected**: Enrich with current title, company details, tenure, and social profiles +- **If knowledge base connected**: Pull the team's screening rubric and hiring manager preferences + +## Tips + +- For best results, provide both the resume AND the job description +- Mention any dealbreakers upfront (e.g., "must be willing to relocate", "no agency candidates") +- If screening multiple candidates for the same role, run this command for each and compare outputs diff --git a/recruiting/commands/source.md b/recruiting/commands/source.md new file mode 100644 index 0000000..77fea1c --- /dev/null +++ b/recruiting/commands/source.md @@ -0,0 +1,82 @@ +--- +description: Build a targeted sourcing strategy for a role +argument-hint: "" +--- + +# Source Candidates + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Build a comprehensive sourcing strategy for a specific role. Produces a prioritized channel plan, Boolean search strings, target company lists, and outreach angles. + +## Input + +The user provides one or more of: +- Job title or role description +- Team and hiring manager context +- Must-have and nice-to-have requirements +- Pasted job description +- Target seniority level or years of experience + +If no input is provided, ask: "What role are you sourcing for? Share the title, key requirements, or paste the full JD." + +## Workflow + +1. **Parse the role** — Extract title, level, core skills, domain, and team context from `$ARGUMENTS` +2. **Define the ideal candidate profile** — Map must-haves vs. nice-to-haves, identify adjacent backgrounds that could transfer +3. **Build Boolean search strings** — Generate 3-5 search strings for LinkedIn, GitHub, Google X-ray, and niche platforms +4. **Identify target companies** — List 10-15 companies likely to employ this talent (competitors, adjacent industries, known talent hubs) +5. **Recommend sourcing channels** — Prioritize channels by expected yield: referrals, LinkedIn, communities, job boards, events, agencies +6. **Draft outreach angles** — Write 2-3 compelling hooks tailored to what would motivate this persona to explore +7. **If ATS connected** — Check for existing candidates in pipeline, silver medalists from past roles, and internal mobility candidates + +## Output Structure + +``` +## Sourcing Strategy: [Role Title] + +### Ideal Candidate Profile +- **Must-haves**: [skills, experience] +- **Nice-to-haves**: [bonus qualifications] +- **Adjacent backgrounds**: [transferable profiles] + +### Boolean Search Strings +1. **LinkedIn**: `[string]` +2. **GitHub/Stack Overflow**: `[string]` +3. **Google X-ray**: `[string]` + +### Target Companies (by tier) +| Tier | Companies | Why | +|------|-----------|-----| +| 1 - Direct competitors | ... | Same domain, likely skills match | +| 2 - Adjacent industry | ... | Transferable skills, different context | +| 3 - Talent hubs | ... | Known for developing this talent | + +### Channel Priority +| Channel | Expected yield | Effort | Recommended actions | +|---------|---------------|--------|-------------------| +| Employee referrals | High | Low | [specific steps] | +| LinkedIn outreach | Medium | Medium | [specific steps] | +| ... | ... | ... | ... | + +### Outreach Angles +**Angle 1**: [hook based on career growth] +**Angle 2**: [hook based on mission/impact] +**Angle 3**: [hook based on technical challenge] + +### Pipeline Check +[If ATS connected: existing candidates, silver medalists, internal mobility] +[If standalone: "Connect your ATS to check for existing candidates"] +``` + +## With Connectors + +- **If ATS connected**: Search for past applicants, silver medalists, and internal candidates before external sourcing +- **If data enrichment connected**: Enrich target company lists with employee counts, growth signals, and funding data +- **If knowledge base connected**: Pull the team's sourcing playbooks, past strategies that worked, and hiring manager preferences + +## Tips + +- Include the hiring manager's LinkedIn profile or team page for better target company identification +- Mention any compensation constraints upfront so channel recommendations account for budget +- For niche roles, specify the exact technical stack or domain expertise needed diff --git a/recruiting/commands/write-jd.md b/recruiting/commands/write-jd.md new file mode 100644 index 0000000..a59c445 --- /dev/null +++ b/recruiting/commands/write-jd.md @@ -0,0 +1,83 @@ +--- +description: Write or improve a job description +argument-hint: "" +--- + +# Write Job Description + +> If you see unfamiliar placeholders or need to check which tools are connected, see [CONNECTORS.md](../CONNECTORS.md). + +Write a compelling, inclusive job description from scratch or improve an existing one. Optimizes for candidate conversion, search discoverability, and legal compliance. + +## Input + +The user provides one or more of: +- Role title and level +- Hiring manager's description of the role +- Existing JD to improve +- Team context, reporting structure +- Must-have vs. nice-to-have requirements +- Compensation range +- Company info or careers page + +If minimal input, ask: "What's the role? Share the title, level, and 2-3 sentences about what this person will do." + +## Workflow + +1. **Define the role** — Extract or clarify title, level, team, reporting structure, and core purpose from `$ARGUMENTS` +2. **Structure the JD** — Organize into standard sections following inclusive writing best practices +3. **Write compelling copy** — Lead with impact and growth, not just requirements. Use active voice, avoid jargon +4. **Optimize requirements** — Separate must-haves from nice-to-haves. Flag requirements that may unnecessarily narrow the pool (years of experience, specific degrees, unnecessary certifications) +5. **Inclusivity review** — Scan for gendered language, exclusionary phrasing, and unnecessary barriers. Apply research-backed inclusive language guidelines +6. **SEO optimization** — Ensure the title and key terms match how candidates actually search +7. **If knowledge base connected** — Pull company voice guidelines, benefits info, and team descriptions + +## Output Structure + +``` +## [Role Title] +**Team**: [Team] | **Reports to**: [Title] | **Level**: [Level] +**Location**: [Location/Remote policy] + +### About the role +[2-3 compelling paragraphs about the role's impact, the team, and why this matters] + +### What you'll do +- [Responsibility framed as impact, not task] +- [Responsibility framed as impact, not task] +- [4-6 total responsibilities] + +### What we're looking for +**Required**: +- [Skill or experience, without arbitrary year counts] +- [3-5 must-haves] + +**Nice to have**: +- [Bonus qualification] +- [2-3 nice-to-haves] + +### What we offer +- [Compensation range if provided] +- [Key benefits and perks] +- [Growth and development opportunities] + +### About [Company] +[Brief company description focused on mission and culture] + +--- +**Inclusivity notes**: [Any flags about language, requirements, or barriers that were adjusted] +**SEO notes**: [Title and keyword optimization suggestions] +``` + +## With Connectors + +- **If knowledge base connected**: Pull company boilerplate, benefits details, team descriptions, and brand voice guidelines +- **If ATS connected**: Check for similar open or past roles, pull the intake form, and post the JD directly +- **If HRIS connected**: Verify the approved headcount, compensation band, and reporting structure + +## Tips + +- Provide the hiring manager's unfiltered description — messy input produces better output than over-polished input +- Mention if this is a backfill vs. new role — it changes the framing +- Include compensation range for significantly better candidate conversion +- For technical roles, specify the actual tech stack rather than generic terms diff --git a/recruiting/skills/candidate-evaluation/SKILL.md b/recruiting/skills/candidate-evaluation/SKILL.md new file mode 100644 index 0000000..30ad3c0 --- /dev/null +++ b/recruiting/skills/candidate-evaluation/SKILL.md @@ -0,0 +1,76 @@ +--- +name: candidate-evaluation +description: "Evaluate candidates against role requirements using structured rubrics and scorecards. Provides consistent, evidence-based assessments with bias mitigation. Trigger with \"evaluate this candidate\", \"score this resume\", \"compare candidates\", \"how does this person stack up\", \"is this a fit\", or when the user shares a resume, candidate profile, or interview feedback and asks for an assessment." +--- + +# Candidate Evaluation + +Structured framework for evaluating candidates consistently across any role. Works standalone with pasted resumes and job descriptions; supercharged when connected to an ATS with scorecards and historical data. + +## How It Works + +- **Standalone**: Paste a resume, LinkedIn summary, or interview notes along with the role requirements. Get a structured evaluation with dimension-by-dimension scoring. +- **With connectors**: Pull candidate data from the ATS, reference the team's rubrics from the knowledge base, and compare against historical hiring decisions. + +## Evaluation Framework + +### Dimension Categories + +Every candidate evaluation maps to these six dimensions. Weight and specific criteria vary by role. + +| Dimension | What to assess | Common signals | +|-----------|---------------|----------------| +| **Technical/functional skills** | Can they do the core work? | Past projects, tools, certifications, portfolio | +| **Domain experience** | Do they know this space? | Industry tenure, relevant companies, market knowledge | +| **Scope and impact** | Have they operated at this level? | Team size led, revenue owned, project complexity | +| **Growth trajectory** | Are they on an upward arc? | Promotion velocity, increasing responsibility, learning signals | +| **Collaboration and communication** | Can they work with the team? | Cross-functional projects, stakeholder management, writing samples | +| **Motivation and alignment** | Will they thrive here? | Career goals, values fit, reason for exploring | + +### Scoring Scale + +| Score | Label | Definition | +|-------|-------|------------| +| 4 | Strong hire | Exceeds the bar — clear evidence across multiple signals | +| 3 | Hire | Meets the bar — sufficient evidence with minor gaps | +| 2 | No hire | Below the bar — significant gaps or insufficient evidence | +| 1 | Strong no hire | Clear misalignment — fundamental gaps or red flags | + +### Bias Mitigation Checklist + +Apply before finalizing any evaluation: + +- Am I evaluating evidence or making assumptions about potential? +- Am I holding all candidates to the same criteria, regardless of background? +- Am I conflating "culture fit" with "similar to people already here"? +- Am I penalizing non-traditional career paths that demonstrate the same competencies? +- Am I over-indexing on pedigree (school, company names) vs. actual accomplishments? +- Would I rate this evidence the same if it came from a different demographic? + +## Comparative Evaluation + +When comparing multiple candidates for the same role: + +1. Score each candidate independently first — never adjust scores after comparing +2. Create a comparison matrix with all dimensions side by side +3. Identify where candidates differentiate (often only 1-2 dimensions matter) +4. Consider team composition — what does the existing team need? +5. Document the reasoning, not just the scores + +## Output Format + +When evaluating a candidate, produce: + +1. **Overall recommendation** with confidence level +2. **Dimension scores** with evidence for each +3. **Key strengths** — top 2-3 with specific proof points +4. **Key risks** — top 2-3 with mitigation strategies or questions to resolve +5. **Comparison** to other candidates if applicable + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| ATS | Pull application history, prior feedback, source attribution | +| Data enrichment | Current title verification, company details, tenure accuracy | +| Knowledge base | Team rubrics, competency frameworks, hiring manager preferences | diff --git a/recruiting/skills/candidate-evaluation/references/red-flags-green-flags.md b/recruiting/skills/candidate-evaluation/references/red-flags-green-flags.md new file mode 100644 index 0000000..0e5e53f --- /dev/null +++ b/recruiting/skills/candidate-evaluation/references/red-flags-green-flags.md @@ -0,0 +1,60 @@ +# Red Flags and Green Flags in Candidate Evaluation + +Patterns to watch for when evaluating candidates. These are signals, not verdicts — always seek additional evidence before making a decision based on a single flag. + +## Green Flags (Positive Signals) + +### Career Trajectory +- **Increasing scope over time** — each role has more responsibility, larger teams, bigger problems +- **Promoted from within** — especially if promoted quickly, signals strong performance +- **Sought out for hard problems** — moved to new teams/projects specifically to fix or build something +- **Startup-to-scale experience** — has seen a company grow through multiple stages +- **Boomerang employee** — left and came back to a former employer (they know what they're getting) + +### Interview Behavior +- **Specific examples with metrics** — "I increased retention by 15% by..." vs. "I improved retention" +- **Acknowledges failures and learnings** — "Here's what went wrong and what I'd do differently" +- **Asks thoughtful questions about the role** — shows they're evaluating fit, not just selling +- **Credits their team** — "We built..." not "I built..." (while still showing their contribution) +- **Demonstrates curiosity** — asks follow-up questions, wants to understand the problem deeply +- **Comfortable saying "I don't know"** — followed by how they'd figure it out + +### References and Reputation +- **Former colleagues want to work with them again** — the ultimate signal +- **Referenced without prompting** — others mention them as a strong collaborator +- **Their previous team members follow them** — people who've worked with them seek them out + +## Red Flags (Concerning Signals) + +### Career Pattern Flags +- **Consistent short tenure** (<18 months repeatedly) — may indicate performance issues, poor judgment, or inability to commit. BUT: normalize for industry (startups have shorter tenure) and early career (first 2-3 roles often shorter) +- **Lateral moves without growth** — moving companies at the same level repeatedly without progression +- **Unexplained gaps** — not that gaps are bad, but evasiveness about them is a flag. Many gaps have great explanations (caregiving, health, travel, education) +- **Title inflation without scope** — "VP" at a 5-person company doesn't equal "VP" at a 5,000-person company + +### Interview Behavior Flags +- **Vague about contributions** — can't articulate what they specifically did vs. what the team did +- **Blames others consistently** — every failure was someone else's fault +- **Can't explain decisions** — "That's just how we did it" without understanding why +- **Oversells and doesn't listen** — treats the interview as a pitch rather than a conversation +- **Different story to different interviewers** — inconsistencies across the interview loop +- **Can't go deep on claimed expertise** — surface-level answers on skills listed as "expert" +- **Negative about every past employer** — one bad experience is normal; a pattern is concerning + +### Reference Check Flags +- **Can't provide peers or reports as references** — only offers managers or friends +- **References are carefully scripted** — feel rehearsed, lack specifics +- **"They'd be great in the right role"** — polite way of saying not a strong performer +- **Hesitation before positive statements** — the pause often says more than the words + +## Flags That Are Not Actually Flags + +Be careful not to penalize candidates for: + +- **Non-traditional backgrounds** — bootcamps, self-taught, career changers may have exactly the skills you need +- **Career gaps** — parenting, health, caregiving, education, travel are all legitimate +- **Accent or communication style** — assess communication effectiveness, not conformity to one style +- **Lack of "name brand" employers** — great talent exists at every company size +- **Being quiet or introverted** — don't confuse confidence with competence +- **Over-qualification** — understand their motivation rather than assuming they'll be bored +- **Job-hopping in the first 5 years of career** — early-career exploration is normal and healthy diff --git a/recruiting/skills/candidate-evaluation/references/scoring-rubrics.md b/recruiting/skills/candidate-evaluation/references/scoring-rubrics.md new file mode 100644 index 0000000..5673e01 --- /dev/null +++ b/recruiting/skills/candidate-evaluation/references/scoring-rubrics.md @@ -0,0 +1,72 @@ +# Scoring Rubrics by Function + +Detailed rubric definitions for the six evaluation dimensions, tailored by function. Use these as starting points and customize for specific roles. + +## Engineering Roles + +### Technical Skills + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Has deep expertise in the core tech stack. Built systems at comparable scale. Can articulate trade-offs and has opinions grounded in experience. Portfolio/contributions demonstrate mastery. | +| 3 - Hire | Solid proficiency in the core stack. Has worked at relevant scale with some guidance. Can learn adjacent technologies quickly. Good problem-solving instincts. | +| 2 - No hire | Surface-level familiarity with the stack. Hasn't worked at the required scale. Struggles to articulate technical decisions or trade-offs. Significant ramp time expected. | +| 1 - Strong no hire | Fundamental gaps in core skills. Misrepresents experience level. Unable to engage in technical discussion at the expected depth. | + +### System Design / Architecture + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Designs for scale, maintainability, and operational concerns unprompted. Considers failure modes. Has built similar systems. Articulates clear reasoning for decisions. | +| 3 - Hire | Produces reasonable designs with appropriate prompting. Considers major trade-offs. May need guidance on edge cases or operational concerns. | +| 2 - No hire | Designs are naive or over-engineered. Misses critical requirements. Doesn't consider scale or failure modes. Needs significant guidance. | +| 1 - Strong no hire | Cannot produce a coherent design. Fundamental misunderstanding of system design principles. | + +### Scope and Impact + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Has owned projects/systems end-to-end with measurable business impact. Led technical strategy for a team or domain. Clear evidence of multiplying others' effectiveness. | +| 3 - Hire | Has completed significant projects independently. Contributed to team strategy. Beginning to mentor others and influence beyond their immediate work. | +| 2 - No hire | Mostly executed defined tasks. Limited evidence of independent project ownership. Scope doesn't match the level we're hiring for. | +| 1 - Strong no hire | Only evidence of task execution with close supervision. Scope is significantly below the target level. | + +## Product Roles + +### Customer Insight + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Deep empathy for users backed by rigorous research methods. Can articulate user needs that users themselves haven't expressed. Has shipped features that meaningfully moved user metrics. | +| 3 - Hire | Good instinct for user needs. Uses data and research to validate hypotheses. Has examples of user-informed product decisions. | +| 2 - No hire | Relies on assumptions about users. Limited evidence of research-informed decisions. May conflate personal preferences with user needs. | +| 1 - Strong no hire | Disconnected from user reality. Makes product decisions without user input. No evidence of user research or data-driven thinking. | + +### Prioritization + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Frameworks-driven prioritization with clear evidence of impact. Can say no convincingly. Balances short-term delivery with long-term strategy. Manages stakeholder expectations masterfully. | +| 3 - Hire | Uses reasonable frameworks to prioritize. Can articulate trade-offs. Generally makes sound bets. May struggle with the most ambiguous situations. | +| 2 - No hire | Prioritizes by recency or loudest voice. Struggles to articulate why one thing matters more than another. Evidence of scope creep or inability to focus. | +| 1 - Strong no hire | No visible prioritization framework. Reactive to requests. Cannot demonstrate making a hard trade-off. | + +## Leadership Roles (Any Function) + +### People Leadership + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Built and developed high-performing teams. Team members have grown significantly under their leadership. Can articulate their management philosophy with specific examples. | +| 3 - Hire | Has managed people effectively. Some evidence of developing others. Generally positive team outcomes. Growing as a leader. | +| 2 - No hire | Limited management experience or mixed results. May rely on authority rather than influence. Team outcomes are unclear or concerning. | +| 1 - Strong no hire | Evidence of poor people management. High team turnover without good explanation. Unable to articulate how they develop others. | + +### Strategic Thinking + +| Score | Signal | +|-------|--------| +| 4 - Strong hire | Sets direction that others follow. Anticipates market shifts. Has made strategic bets that paid off. Thinks in systems, not just features. | +| 3 - Hire | Contributes meaningfully to strategy. Can connect daily work to broader goals. Shows evidence of forward-looking thinking. | +| 2 - No hire | Primarily tactical. Struggles to articulate a strategic perspective. May execute well but doesn't set direction. | +| 1 - Strong no hire | No strategic capability. Purely reactive. Cannot think beyond the current quarter. | diff --git a/recruiting/skills/diversity-recruiting/SKILL.md b/recruiting/skills/diversity-recruiting/SKILL.md new file mode 100644 index 0000000..d638729 --- /dev/null +++ b/recruiting/skills/diversity-recruiting/SKILL.md @@ -0,0 +1,105 @@ +--- +name: diversity-recruiting +description: "Apply inclusive hiring practices, mitigate bias in recruiting processes, and diversify sourcing channels. Covers inclusive JD writing, structured evaluation, diverse pipelines, and equitable processes. Trigger with \"diverse candidates\", \"inclusive hiring\", \"reduce bias\", \"DEI recruiting\", \"diverse pipeline\", \"inclusive job description\", or when the user discusses diversity goals, representation, bias concerns, or equitable hiring practices." +--- + +# Diversity Recruiting + +Evidence-based practices for building inclusive hiring processes that reach diverse talent pools and evaluate candidates equitably. Not a separate workflow — these principles integrate into every stage of recruiting. + +## How It Works + +- **Standalone**: Get actionable guidance on making any part of your hiring process more inclusive, from JD writing to offer negotiations. +- **With connectors**: Analyze pipeline diversity metrics from the ATS, audit JDs from the knowledge base, and track representation goals. + +## Inclusive Practices by Stage + +### 1. Job Description & Requirements + +**Common issues and fixes:** + +| Problem | Why it matters | Fix | +|---------|---------------|-----| +| Gendered language ("rockstar", "ninja", "aggressive") | Research shows women are less likely to apply | Use neutral alternatives ("skilled", "driven", "tenacious") | +| Excessive requirements | Women apply when meeting 100% of requirements; men at 60% | Separate must-haves from nice-to-haves, reduce total requirements | +| Unnecessary degree requirements | Excludes self-taught and non-traditional backgrounds | List specific skills needed instead | +| Years of experience requirements | Disadvantages career changers and returners | Describe proficiency level needed, not years | +| "Culture fit" language | Signals homogeneity | Use "culture add" — what new perspective does this person bring? | +| No salary range | Perpetuates pay gaps for candidates with less negotiation leverage | Include the range | + +**Language audit checklist:** +- Run the JD through a gender decoder tool (or just search for: aggressive, competitive, dominant, challenging → replace with: collaborative, supportive, analytical, innovative) +- Remove "preferred" qualifications that aren't actually preferred +- Add an explicit inclusion statement that's specific, not generic +- Ensure physical requirements are truly job-related + +### 2. Sourcing + +**Expand beyond the usual channels:** +- HBCUs, HSIs, and tribal colleges +- Historically underrepresented professional organizations (NSBE, SHPE, AnitaB, /dev/color, Lesbians Who Tech, Out in Tech, Disability:IN) +- Returnship programs and career re-entry networks +- Coding bootcamps and alternative education programs +- Community colleges and vocational programs +- Veteran transition programs +- International talent pools + +**Referral program audit:** +- Referrals tend to replicate existing demographics — this isn't malicious, it's network effects +- Supplement referrals with structured outbound sourcing to diverse communities +- Consider bonus incentives specifically for diverse referrals (legal in most jurisdictions — consult your legal team) + +### 3. Screening + +- Use structured screening criteria defined before reviewing candidates +- Blind resume review where possible (remove names, photos, schools) +- Evaluate accomplishments, not credentials +- Watch for "pattern matching" to existing team members + +### 4. Interviewing + +- **Structured interviews** are the single most effective bias reducer +- Same questions for every candidate at a given stage +- Pre-defined rubrics with specific behavioral anchors +- Diverse interview panels where possible +- Train interviewers on common biases: + - **Affinity bias**: Preferring people similar to yourself + - **Halo/horn effect**: One strong/weak signal coloring the whole evaluation + - **Confirmation bias**: Seeking evidence to confirm initial impression + - **Attribution bias**: Attributing success to environment for some, to skill for others + +### 5. Offer & Negotiation + +- Present the same offer structure to all candidates at the same level +- Don't ask for salary history (illegal in many jurisdictions, perpetuates gaps) +- Have a defined negotiation framework — don't give more to those who negotiate harder +- Be transparent about the comp philosophy and how the offer was built + +## Metrics to Track + +| Metric | What it measures | How to use it | +|--------|-----------------|---------------| +| **Pipeline diversity by stage** | Where diverse candidates drop off | Identify the leaky stage and investigate | +| **Source diversity** | Which channels produce diverse candidates | Invest more in high-diversity channels | +| **Interview pass rate by demographic** | Whether evaluation is equitable | Flag disparities for investigation | +| **Offer rate parity** | Whether offers are extended equitably | Audit decision-making if disparities exist | +| **Acceptance rate parity** | Whether offers are equally attractive | Check comp equity and candidate experience | +| **Hiring manager diversity** | Pattern of who gets hired onto which teams | Coach managers with persistent gaps | + +## Output Format + +When advising on inclusive hiring, produce: + +1. **Assessment of current process** — where bias is most likely to enter +2. **Specific recommendations** — actionable changes, not generic advice +3. **Language suggestions** — concrete rewrites, not just "be more inclusive" +4. **Sourcing channels** — specific communities and organizations relevant to the role +5. **Metrics to track** — what to measure and what to watch for + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| ATS | Pipeline diversity analytics, stage-by-stage demographic data | +| Knowledge base | Company DEI commitments, ERG contacts, approved language guidelines | +| Data enrichment | Identify diverse professional communities and organizations | diff --git a/recruiting/skills/diversity-recruiting/references/inclusive-language-guide.md b/recruiting/skills/diversity-recruiting/references/inclusive-language-guide.md new file mode 100644 index 0000000..293cc3b --- /dev/null +++ b/recruiting/skills/diversity-recruiting/references/inclusive-language-guide.md @@ -0,0 +1,107 @@ +# Inclusive Language Guide for Recruiting + +Specific language substitutions and patterns for writing inclusive job descriptions, outreach messages, and candidate communications. + +## Gendered Language Replacements + +| Instead of | Use | Why | +|-----------|-----|-----| +| Rockstar, ninja, guru | Expert, specialist, skilled professional | Coded masculine, can feel exclusionary | +| Aggressive | Ambitious, driven, tenacious | Research links "aggressive" to masculine stereotypes | +| Competitive | Results-oriented, high-performing | Can discourage collaborative candidates | +| Dominant | Leading, influential, authoritative | Aggressive connotation | +| Manpower | Workforce, staffing, team capacity | Gendered term | +| Chairman | Chair, chairperson | Gendered term | +| Guys | Team, folks, everyone, y'all | Gendered when addressing groups | +| He/she | They (singular) | More inclusive of all gender identities | +| Man-hours | Person-hours, work hours | Gendered term | + +## Exclusionary Requirement Patterns + +### Degree Requirements + +| Instead of | Use | Why | +|-----------|-----|-----| +| "Bachelor's degree required" | "Bachelor's degree or equivalent experience" | Opens the door to self-taught and alternative education | +| "CS degree from a top university" | "Strong computer science fundamentals" | Screens on knowledge, not pedigree | +| "MBA preferred" | "Experience in [specific business skill]" | Names the actual skill needed | + +### Experience Requirements + +| Instead of | Use | Why | +|-----------|-----|-----| +| "10+ years of experience" | "Deep expertise in [skill], demonstrated through complex projects" | Years ≠ expertise; penalizes career changers | +| "5+ years in Python" | "Proficient in Python with production experience" | Tests skill, not time | +| "Native English speaker" | "Strong written and verbal communication skills" | Legal risk, excludes bilingual talent | +| "Digital native" | "Comfortable with [specific tools/platforms]" | Age-coded language | + +### Culture Fit Language + +| Instead of | Use | Why | +|-----------|-----|-----| +| "Culture fit" | "Culture add — what new perspective do you bring?" | Fit ≠ sameness; add implies growth | +| "Work hard, play hard" | "We're a team that takes both our work and well-being seriously" | Signals overwork, can exclude parents/caregivers | +| "Fast-paced environment" | "We ship frequently and iterate based on feedback" | Be specific about what "fast" means | +| "Young and dynamic team" | "Our team brings diverse experience and energy" | Age-discriminatory | +| "Recent graduate" | "Early-career professional" | Excludes non-traditional timelines | + +## Inclusive Job Description Checklist + +- [ ] Does the JD focus on what the person will DO, not who they ARE? +- [ ] Are must-haves truly must-haves (required on day 1)? +- [ ] Is the nice-to-have list short (3 or fewer)? +- [ ] Are there fewer than 10 total requirements? +- [ ] Is a salary range included? +- [ ] Is the inclusion statement specific (not just "we're an equal opportunity employer")? +- [ ] Are physical requirements only listed if genuinely job-related? +- [ ] Is the language free of jargon that insiders would understand but outsiders wouldn't? +- [ ] Would someone from a non-traditional background see themselves in this JD? +- [ ] Is the application process accessible (screen reader compatible, reasonable time expectations)? + +## Inclusive Outreach Language + +### Opening Lines + +**Less inclusive:** +> "I came across your profile and your background is impressive." + +**More inclusive:** +> "Your work on [specific project] caught my attention because [specific reason related to the role]." + +Why: The first is generic and could trigger imposter syndrome. The second validates specific work. + +### Describing the Opportunity + +**Less inclusive:** +> "We're looking for a brilliant engineer who can hit the ground running." + +**More inclusive:** +> "We're building [specific thing] and looking for someone who's excited about [specific challenge]. You'd have [specific support — mentoring, onboarding, team structure] to ramp up." + +Why: "Brilliant" and "hit the ground running" signal that only already-perfect candidates should apply. The second version signals support and growth. + +### Closing + +**Less inclusive:** +> "Let me know if you're interested. We move fast so the sooner the better." + +**More inclusive:** +> "Would you be open to a 20-minute conversation to explore this? Happy to work around your schedule." + +Why: Low-pressure, specific ask, respects their time. + +## Inclusive Rejection Language + +**Template:** +> Thank you for taking the time to interview with us for the [Role] position. After careful consideration, we've decided to move forward with another candidate whose experience more closely aligns with our current needs. +> +> This was a competitive process, and your [specific positive — skills, experience, approach] made a strong impression. We'd welcome the opportunity to stay in touch for future roles that might be a better match. +> +> If you'd find it helpful, I'm happy to share more specific feedback about our decision. + +**Key principles:** +- Be timely (within 48 hours of decision) +- Be specific about what was strong (not just "you were great") +- Offer feedback (not everyone wants it, but offer) +- Leave the door open genuinely +- Never ghost — it's the single biggest employer brand killer diff --git a/recruiting/skills/employer-branding/SKILL.md b/recruiting/skills/employer-branding/SKILL.md new file mode 100644 index 0000000..4d77440 --- /dev/null +++ b/recruiting/skills/employer-branding/SKILL.md @@ -0,0 +1,101 @@ +--- +name: employer-branding +description: "Craft employer brand messaging, EVP content, job marketing copy, and candidate experience communications. Helps recruiting teams tell a compelling story about why candidates should join. Trigger with \"employer brand\", \"EVP\", \"why join us\", \"careers page\", \"job marketing\", \"candidate experience\", \"recruitment marketing\", or when the user discusses attracting candidates, company culture messaging, or talent brand differentiation." +--- + +# Employer Branding + +Framework for crafting authentic, compelling employer brand content that attracts the right candidates. Covers EVP development, job marketing, candidate communications, and experience design. + +## How It Works + +- **Standalone**: Describe your company, team, or role and get employer brand messaging, recruitment marketing copy, and candidate communication templates. +- **With connectors**: Pull existing brand materials from the knowledge base, analyze competitor positioning via web search, and distribute content through chat and email. + +## Employer Value Proposition (EVP) Framework + +### EVP Pillars + +Every compelling EVP addresses these five dimensions: + +| Pillar | Questions to answer | Examples | +|--------|-------------------|---------| +| **Mission and impact** | What problem are we solving? Why does it matter? | "We're making healthcare accessible to 100M people" | +| **Growth and development** | How will people grow here? What will they learn? | "Engineers ship to production in week 1, lead projects by month 3" | +| **People and culture** | Who will they work with? What's the working style? | "Small teams, high autonomy, weekly demo culture" | +| **Compensation and rewards** | How do we invest in our people? | "Top-quartile comp, meaningful equity, unlimited learning budget" | +| **Work environment** | How, when, and where do people work? | "Remote-first, async-heavy, quarterly offsites" | + +### EVP Development Process + +1. **Audit current messaging** — Careers page, JDs, Glassdoor profile, social media +2. **Identify real differentiators** — What's genuinely different, not just "we have great people" +3. **Talk to recent hires** — Why did they actually join? What surprised them? +4. **Talk to candidates who declined** — Why did they choose elsewhere? +5. **Map to talent segments** — Different audiences value different pillars + +### Messaging Principles + +- **Specific > Generic**: "We deploy 50 times a day" beats "We move fast" +- **Evidence > Claims**: Show, don't tell — numbers, stories, examples +- **Honest > Aspirational**: Don't sell a culture you don't have +- **Candidate-centric**: Frame everything as "what's in it for you", not "what we need" +- **Differentiated**: If a competitor could say the same thing, it's not a differentiator + +## Content Templates + +### Job Marketing Copy (Social/LinkedIn) + +``` +[Hook — specific, interesting detail about the role or team] + +We're looking for [role] to [specific impact they'll have]. + +What makes this different: +→ [Differentiator 1 — specific] +→ [Differentiator 2 — specific] +→ [Differentiator 3 — specific] + +[Link] | [Compensation range if applicable] +``` + +### Recruiter Outreach Hook Library + +| Angle | Template | Best for | +|-------|----------|----------| +| **Mission** | "We're [specific mission statement]. I'm reaching out because [their background] suggests you care about [related problem]..." | Mission-driven candidates | +| **Technical challenge** | "[Specific technical problem] — your work on [their project] caught my eye because..." | Engineers, technical roles | +| **Growth** | "We're [stage/growth signal] and building [team]. This is a [specific growth opportunity]..." | Career builders | +| **People** | "I work with [notable person/team detail]. We're looking for someone who [specific trait they demonstrate]..." | Culture-motivated candidates | + +### Candidate Experience Communications + +Key touchpoints that shape employer brand: + +| Touchpoint | Timing | Must include | +|-----------|--------|-------------| +| **Application confirmation** | Immediate | Timeline, what to expect, human touch | +| **Status update** | Every 5-7 business days | Where they are, next step, honest timeline | +| **Interview prep** | 48 hours before | Who they'll meet, format, how to prepare | +| **Post-interview** | Within 24 hours | Thank you, timeline for next steps | +| **Rejection** | Within 48 hours of decision | Specific feedback (if possible), warmth, future openings | +| **Offer** | Within 2 days of debrief | Excitement, full package details, decision support | + +## Output Format + +When creating employer brand content, produce: + +1. **Audience analysis** — who we're targeting and what they value +2. **Key messages** — 3-5 specific, differentiated value propositions +3. **Content** — ready-to-use copy for the requested channel/format +4. **Tone guidance** — how the content should feel and what to avoid +5. **Distribution plan** — where and how to share the content + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| Knowledge base | Existing brand guidelines, culture docs, benefits information | +| Chat | Distribute content to hiring channels, collect employee testimonials | +| Email | Send candidate communications, nurture campaigns | +| Cloud storage | Access brand assets, photos, templates | diff --git a/recruiting/skills/interview-design/SKILL.md b/recruiting/skills/interview-design/SKILL.md new file mode 100644 index 0000000..cff4860 --- /dev/null +++ b/recruiting/skills/interview-design/SKILL.md @@ -0,0 +1,84 @@ +--- +name: interview-design +description: "Design structured interview processes with competency coverage, question banks, and interviewer assignments. Ensures every competency is assessed without redundancy. Trigger with \"design an interview loop\", \"set up interviews for this role\", \"what should we ask\", \"interview plan\", \"who should interview\", or when the user discusses structuring an interview process for a role." +--- + +# Interview Design + +Framework for building structured interview loops that comprehensively assess candidates while respecting everyone's time. Ensures competency coverage across stages with no gaps or redundancy. + +## How It Works + +- **Standalone**: Describe the role and get a complete interview plan with stages, competencies, questions, and interviewer profiles. +- **With connectors**: Pull the existing interview plan from the ATS, check interviewer availability on the calendar, and reference the company's competency framework from the knowledge base. + +## Interview Loop Architecture + +### Standard Stages + +| Stage | Purpose | Duration | Who | Competencies | +|-------|---------|----------|-----|-------------| +| **Recruiter screen** | Qualification, motivation, logistics | 30 min | Recruiter | Motivation, baseline qualification, logistics | +| **Hiring manager screen** | Role fit, scope, working style | 45 min | Hiring manager | Domain knowledge, scope/impact, collaboration | +| **Technical/skills assessment** | Core functional skills | 60 min | Technical peer | Technical skills, problem-solving, depth | +| **Cross-functional/collaboration** | Teamwork, communication, influence | 45 min | Cross-functional partner | Communication, stakeholder management, influence | +| **Culture and values** | Values alignment, growth mindset | 30-45 min | Culture interviewer | Values alignment, self-awareness, growth | +| **Executive/bar raiser** | Overall bar, strategic thinking | 30-45 min | Skip-level leader | Strategic thinking, judgment, leadership | + +### Competency Coverage Matrix + +When designing a loop, build a matrix ensuring every critical competency is assessed at least once, and the most important ones are assessed twice by different interviewers: + +``` + Recruiter HM Screen Technical Collab Culture Exec +Technical skills — ○ ● — — ○ +Domain expertise ○ ● ○ — — — +Problem-solving — ○ ● ○ — — +Communication ○ ○ — ● ○ — +Leadership/influence — ○ — ● — ● +Values alignment ○ — — — ● ○ +Growth/learning — — ○ — ● — +Strategic thinking — ○ — — — ● + +● = primary assessor ○ = secondary signal — = not assessed +``` + +### Question Design Principles + +1. **Behavioral questions** — "Tell me about a time when..." with STAR follow-ups +2. **Situational questions** — "How would you approach..." for judgment and thinking style +3. **Technical questions** — Role-specific skills demonstration +4. **Case questions** — Real problems from the team (anonymized) to see their approach + +For each question, define: +- The competency being assessed +- What "great" looks like (specific, observable behaviors) +- What "concerning" looks like +- Follow-up probes to go deeper + +### Calibration Guide + +Before the first candidate, align the interview panel: +- Review the role requirements and hiring bar together +- Walk through sample answers at each scoring level +- Clarify what "meets the bar" means for this specific role and level +- Assign specific competencies to each interviewer — no overlap except intentional double-coverage + +## Output Format + +When designing an interview loop, produce: + +1. **Loop summary** — stages, interviewers, total candidate time +2. **Competency coverage matrix** — visual map of who assesses what +3. **Per-stage interview guide** — questions, rubrics, time allocation +4. **Calibration agenda** — what to align on before interviews start +5. **Logistics checklist** — scheduling, tools, candidate communication + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| ATS | Pull existing interview plans, submit scorecards, configure stages | +| Calendar | Check interviewer availability, schedule the loop, send invites | +| Knowledge base | Access competency frameworks, question banks, past interview guides | +| Video | Set up interview rooms and meeting links | diff --git a/recruiting/skills/interview-design/references/interview-formats.md b/recruiting/skills/interview-design/references/interview-formats.md new file mode 100644 index 0000000..81ee433 --- /dev/null +++ b/recruiting/skills/interview-design/references/interview-formats.md @@ -0,0 +1,107 @@ +# Interview Format Guide + +Different interview formats and when to use each one. Choose formats based on what you're assessing, the role level, and practical constraints. + +## Format Comparison + +| Format | Best for | Duration | Assesses | Candidate experience | +|--------|----------|----------|----------|---------------------| +| **Behavioral** | All roles | 30-60 min | Past behavior, judgment, values | Comfortable (familiar format) | +| **Technical deep-dive** | Engineering, data, technical roles | 45-90 min | Technical depth, problem-solving | Can be stressful — set expectations | +| **Case study** | Product, strategy, consulting roles | 45-60 min | Analytical thinking, frameworks | Engaging if well-designed | +| **Live coding** | Engineering (IC) | 45-60 min | Coding ability, communication | High variance — standardize carefully | +| **Take-home project** | Engineering, design | 2-4 hours | Quality of work, process | Respectful of time if well-scoped | +| **Portfolio review** | Design, marketing, content | 45-60 min | Craft quality, decision-making | Natural and conversational | +| **Panel interview** | All roles (senior+) | 45-60 min | Multiple perspectives at once | Can be intimidating — keep panel small | +| **Working session** | All roles | 60-90 min | Collaboration, real-world skills | Highly authentic, shows working style | +| **Reverse interview** | Senior+ candidates | 30 min | Candidate's judgment and priorities | Candidates appreciate this | + +## Format Deep Dives + +### Behavioral Interview + +**Structure:** +1. Opening and rapport (5 min) +2. Behavioral questions (35-45 min, 3-5 questions) +3. Candidate questions (10 min) +4. Close and next steps (5 min) + +**Tips:** +- Use the STAR follow-up pattern: Situation → Task → Action → Result +- If the candidate is vague, probe: "What specifically did *you* do?" +- Listen for "we" vs. "I" — both are fine, but understand their specific contribution +- Allow silence — don't rush to fill it + +### Technical Deep-Dive + +**Structure:** +1. Context setting (5 min) — "We're going to explore [topic]. I'll start with some foundational questions and go deeper." +2. Breadth questions (15 min) — Assess range of knowledge +3. Depth questions (25-35 min) — Drill into areas of claimed expertise +4. Architecture/design (15-20 min) — Apply knowledge to a realistic problem +5. Candidate questions (5-10 min) + +**Tips:** +- Start broad and narrow based on their answers — let the candidate lead you to their strengths +- "I don't know" is a valid answer — follow up with "how would you find out?" +- Don't assess memorization — assess understanding and reasoning +- Give partial credit generously; assess the approach, not just the answer + +### Take-Home Project + +**Design principles:** +- **Time-boxed**: Clearly state "this should take 2-4 hours" and mean it +- **Realistic**: Based on actual work they'd do in the role +- **Open-ended enough**: Multiple valid approaches, room for creativity +- **Paid**: If longer than 2 hours, compensate the candidate's time +- **Clear evaluation criteria**: Share them upfront so candidates know what matters + +**What to evaluate:** +- Code quality and readability (not cleverness) +- Approach and architecture decisions +- Testing and edge case handling +- Documentation and communication +- How they handle ambiguity in the spec + +**Follow-up session (30-45 min):** +- Walk through their solution +- Ask about trade-offs and alternatives considered +- Discuss how they'd extend or scale it +- Probe decisions that seem unusual + +### Working Session + +**Structure:** +1. Frame the problem (10 min) — share a real (anonymized) challenge the team is facing +2. Collaborative work (40-60 min) — work on it together, as if they were already on the team +3. Debrief (10-15 min) — discuss the experience, what they'd do with more time + +**Why it works:** +- Shows how they think and collaborate in real-time +- Reduces performance anxiety — feels like work, not a test +- Gives the candidate a realistic preview of the job +- Hard to fake — you see their actual working style + +**Tips:** +- Choose a problem that's genuinely unsolved (they might teach you something) +- Participate as a peer, not an evaluator — make it collaborative +- Note how they ask clarifying questions, handle disagreement, and iterate + +## Scheduling Considerations + +### Full Loop Timeline + +| Role level | Total interview time | Stages | Typical timeline | +|-----------|---------------------|--------|-----------------| +| Junior/Mid | 3-4 hours | 3-4 stages | 1-2 weeks | +| Senior | 4-5 hours | 4-5 stages | 2-3 weeks | +| Staff/Lead | 5-6 hours | 5-6 stages | 2-4 weeks | +| Director+ | 6-8 hours | 5-7 stages | 3-5 weeks | + +### Reducing Candidate Burden + +- Combine stages into a half-day or full-day onsite when possible +- Virtual interviews: max 4 hours in a day with breaks +- Don't repeat questions across stages — coordinate the panel +- Respect that candidates have day jobs — offer flexible scheduling +- Move fast — every extra day in the process is a chance to lose the candidate diff --git a/recruiting/skills/interview-design/references/question-bank.md b/recruiting/skills/interview-design/references/question-bank.md new file mode 100644 index 0000000..19f7173 --- /dev/null +++ b/recruiting/skills/interview-design/references/question-bank.md @@ -0,0 +1,110 @@ +# Interview Question Bank + +Structured questions organized by competency. Each question includes the competency assessed, follow-up probes, and what to listen for. Customize for specific roles and levels. + +## Behavioral Questions (STAR Format) + +### Problem Solving and Technical Judgment + +**Q: Tell me about a time you had to make a technical decision with incomplete information. What was the situation, and how did you approach it?** +- Follow-up: What information did you wish you had? How did you mitigate the risk of being wrong? +- Strong signal: Structured approach to ambiguity, considered reversibility, gathered what data they could, made a call and monitored results +- Weak signal: Waited for perfect information, made a gut call without reasoning, doesn't acknowledge the uncertainty + +**Q: Describe a project where you had to significantly change your approach mid-stream. What triggered the change and how did you handle it?** +- Follow-up: How did you communicate the change to stakeholders? What would you do differently? +- Strong signal: Recognized signals early, adapted without ego, communicated proactively, salvaged what they could +- Weak signal: Blamed external factors, didn't learn from the experience, rigid thinking + +**Q: Walk me through the most complex system/project/campaign you've worked on. What made it complex, and how did you manage that complexity?** +- Follow-up: What would you simplify if you could do it again? Where did complexity bite you? +- Strong signal: Can articulate what made it complex (not just "it was big"), has a framework for managing complexity, identifies unnecessary complexity +- Weak signal: Equates size with complexity, no framework, seems overwhelmed recounting it + +### Collaboration and Communication + +**Q: Tell me about a time you disagreed with a colleague or manager on an important decision. How did you handle it?** +- Follow-up: What was the outcome? Would you approach it differently now? +- Strong signal: Raised disagreement constructively, supported data over opinion, committed to the decision once made (even if they disagreed), maintained the relationship +- Weak signal: Avoided the conflict, escalated without trying to resolve, held a grudge, "I was right and they were wrong" + +**Q: Describe a time you had to influence someone without authority. What was your approach?** +- Follow-up: What was at stake? How did you read the other person's priorities? +- Strong signal: Understood the other person's motivations, framed their case in terms of shared goals, was persistent but respectful +- Weak signal: Used pressure, went around them, gave up quickly, doesn't distinguish influence from authority + +**Q: Tell me about a time you received tough feedback. How did you respond?** +- Follow-up: What changed as a result? How do you typically seek feedback? +- Strong signal: Listened without defensiveness, took concrete action, sought more feedback after, can describe what they learned +- Weak signal: Dismisses the feedback, blames the messenger, no evidence of change, hasn't received feedback (unlikely — more likely they don't recall or can't be vulnerable) + +### Leadership and Ownership + +**Q: Tell me about a time you took ownership of something outside your formal responsibilities. Why, and what happened?** +- Follow-up: How did others respond? Did it create any friction? +- Strong signal: Saw a gap and filled it proactively, navigated ambiguity about ownership, created value, handled any political dynamics maturely +- Weak signal: Can't think of an example, only does what's assigned, waited to be asked + +**Q: Describe a situation where you had to get a group of people aligned around a direction they weren't initially bought into.** +- Follow-up: What resistance did you encounter? How did you handle dissenters? +- Strong signal: Understood individual concerns, built buy-in incrementally, used data and narrative, allowed space for dissent while moving forward +- Weak signal: Used authority, ignored concerns, achieved compliance not alignment + +**Q: Tell me about a time you made a mistake that had real consequences. What happened and what did you do?** +- Follow-up: How did you prevent it from happening again? What did you learn? +- Strong signal: Takes full ownership, describes the impact honestly, took action to fix and prevent, shares the learning willingly +- Weak signal: Minimizes the mistake, blames circumstances, no evidence of change, struggles to recall any mistakes (low self-awareness) + +### Growth and Learning + +**Q: What's something you've deliberately gotten better at in the last year? How did you approach improving?** +- Follow-up: How do you know you've improved? What's next on your development list? +- Strong signal: Specific skill, deliberate practice, can measure progress, shows self-awareness about gaps +- Weak signal: Vague ("I'm always learning"), can't point to specific improvement, no method + +**Q: Describe a time you had to learn something quickly to be effective in your role. How did you ramp up?** +- Follow-up: How long did it take? What resources did you use? How did you know you were ready? +- Strong signal: Structured approach to learning, leveraged multiple resources, asked for help, applied knowledge quickly +- Weak signal: Passive learning, took too long, didn't ask for help, can't describe their method + +## Situational Questions + +**Q: Imagine you're starting this role and in your first week you discover that [specific scenario relevant to the role]. How would you approach it?** + +**Q: You have three priorities, all due this week, and your manager adds a fourth. How do you handle it?** + +**Q: A teammate consistently delivers late, impacting your work. Your manager hasn't noticed. What do you do?** + +**Q: You've proposed a solution that the team disagrees with, but you believe strongly it's the right approach. What's your next move?** + +## Role-Specific Starters + +### Engineering +- "Walk me through how you'd design [system relevant to the role]" +- "Here's a bug report — walk me through how you'd investigate and fix it" +- "How do you decide what to test and at what level?" +- "Tell me about a time you improved the performance or reliability of a system" + +### Product Management +- "How would you prioritize these five feature requests?" (provide realistic examples) +- "Walk me through how you'd measure success for [feature]" +- "A key metric dropped 10% this week. Walk me through your investigation" +- "How do you decide when to kill a feature?" + +### Design +- "Walk me through your design process for [relevant project type]" +- "How do you handle design feedback you disagree with?" +- "Show me a design you're proud of and one you'd do differently" +- "How do you balance user needs with business goals?" + +### Sales +- "Walk me through your discovery process with a new prospect" +- "Tell me about a deal you lost. What happened and what did you learn?" +- "How do you handle a price objection when you know your product is more expensive?" +- "Describe your approach to managing a complex, multi-stakeholder deal" + +### Leadership +- "How do you assess and develop your team members?" +- "Tell me about a time you had to let someone go. How did you approach it?" +- "How do you set goals for your team and hold them accountable?" +- "Describe how you've built a team from scratch" diff --git a/recruiting/skills/job-architecture/SKILL.md b/recruiting/skills/job-architecture/SKILL.md new file mode 100644 index 0000000..e42c35b --- /dev/null +++ b/recruiting/skills/job-architecture/SKILL.md @@ -0,0 +1,71 @@ +--- +name: job-architecture +description: "Define job levels, competency frameworks, and role scoping for recruiting. Helps clarify what a role actually needs versus what's traditionally listed. Trigger with \"what level is this\", \"define this role\", \"competency framework\", \"leveling guide\", \"what should the requirements be\", or when the user discusses role definition, leveling, job families, or career ladders in a hiring context." +--- + +# Job Architecture + +Framework for defining roles, levels, and competencies in the context of hiring. Ensures job requirements reflect actual needs rather than inflated wish lists, and that leveling is consistent across the organization. + +## How It Works + +- **Standalone**: Describe a role and get a structured breakdown of level-appropriate scope, competencies, and requirements. +- **With connectors**: Pull existing leveling frameworks from the knowledge base, check HRIS for approved headcount and band, and reference similar roles in the ATS. + +## Leveling Framework + +### Generic Engineering Levels (adapt to function) + +| Level | Title pattern | Scope | Experience signal | Key differentiators | +|-------|--------------|-------|-------------------|-------------------| +| L3 | Junior/Associate | Defined tasks with guidance | 0-2 years | Learning, executing, asking good questions | +| L4 | Mid-level | Features and projects with direction | 2-5 years | Independent execution, some ambiguity tolerance | +| L5 | Senior | Projects end-to-end, mentors others | 5-8 years | Owns outcomes, influences design, unblocks others | +| L6 | Staff/Lead | Multi-project, cross-team impact | 8-12 years | Sets direction, navigates org complexity, multiplies team | +| L7 | Principal/Director | Org-wide strategy and execution | 12+ years | Defines the roadmap, industry perspective, builds teams | + +### Competency Dimensions by Function + +**Engineering**: Technical depth, system design, code quality, debugging, architecture, mentorship +**Product**: Customer insight, prioritization, stakeholder management, data fluency, strategy +**Design**: Craft quality, user research, systems thinking, cross-functional collaboration, critique +**Sales**: Pipeline generation, discovery, negotiation, relationship building, forecasting +**Marketing**: Positioning, channel strategy, measurement, creative direction, brand stewardship +**Operations**: Process design, automation, vendor management, compliance, scalability + +### Requirements Calibration + +When defining role requirements, apply these filters: + +1. **Is this truly required on day 1?** If they can learn it in 3 months, it's a nice-to-have +2. **Does this requirement serve the work or the hiring manager's preference?** Remove vanity requirements +3. **Does this artificially narrow the pool?** "10 years of React" excludes excellent engineers who used other frameworks +4. **Can we test this in the interview instead of screening for it?** Skills > credentials +5. **Are we using years of experience as a proxy for something measurable?** Name the actual skill + +### Role Scoping Checklist + +When scoping a new role: +- What problem does this hire solve? (not "what will they do" but "what changes when they're here") +- What does this person own that nobody currently owns? +- What does success look like at 30/60/90 days? +- What's the minimum viable version of this role? +- Could an internal move or contractor solve this instead? + +## Output Format + +When defining a role or leveling, produce: + +1. **Role summary** — purpose, scope, team context +2. **Level assessment** — where this role sits in the framework and why +3. **Competency profile** — weighted dimensions for this specific role +4. **Requirements** — must-haves vs. nice-to-haves, calibrated against the filters above +5. **Success metrics** — what "great" looks like at 30/60/90 days + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| Knowledge base | Pull existing leveling guides, career ladders, and competency frameworks | +| HRIS | Check approved headcount, compensation bands, and org structure | +| ATS | Reference similar roles, past job descriptions, and hiring outcomes | diff --git a/recruiting/skills/job-architecture/references/leveling-frameworks.md b/recruiting/skills/job-architecture/references/leveling-frameworks.md new file mode 100644 index 0000000..0f37402 --- /dev/null +++ b/recruiting/skills/job-architecture/references/leveling-frameworks.md @@ -0,0 +1,78 @@ +# Leveling Frameworks by Function + +Reference frameworks for calibrating role levels across functions. These are generic starting points — every company should customize to their own context and title conventions. + +## Engineering + +| Level | Titles | Scope | Technical | Leadership | Impact | +|-------|--------|-------|-----------|-----------|--------| +| **IC1** | Junior Engineer, Associate | Tasks and small features | Developing core skills, learning the codebase | None expected | Individual tasks | +| **IC2** | Software Engineer | Features within a domain | Solid in primary stack, learning adjacent areas | Helps onboard new hires | Features and bug fixes | +| **IC3** | Senior Engineer | Projects end-to-end | Deep in primary stack, breadth across team's systems | Mentors IC1-2, reviews code, influences design | Projects with team impact | +| **IC4** | Staff Engineer | Multi-project, cross-team | Expert in domain, sets technical standards | Leads technical direction, unblocks others, sponsors initiatives | Cross-team impact | +| **IC5** | Principal Engineer | Org-wide | Industry-recognized expertise | Shapes org-wide technical strategy, builds engineering culture | Org-wide impact | +| **M1** | Engineering Manager | Single team (5-10) | Enough to evaluate and guide | Hires, develops, performance manages team | Team outcomes | +| **M2** | Senior EM / Director | Multiple teams | Broad technical understanding | Builds and leads managers, org design | Multi-team outcomes | +| **M3** | VP Engineering | Engineering org | Strategic technical judgment | Shapes engineering culture, executive leadership | Org outcomes | + +## Product Management + +| Level | Titles | Scope | Skill focus | Impact | +|-------|--------|-------|------------|--------| +| **IC1** | Associate PM | Single feature area | Execution, data analysis, customer empathy | Feature-level improvements | +| **IC2** | Product Manager | Product area / surface | Strategy + execution, stakeholder management | Product area growth | +| **IC3** | Senior PM | Product line | Cross-functional leadership, market insight, vision | Product line outcomes | +| **IC4** | Group PM / Principal | Multiple product lines | Portfolio strategy, organizational influence | Cross-product impact | +| **IC5** | VP Product | Product org | Business strategy, executive leadership | Business outcomes | + +## Design + +| Level | Titles | Scope | Skill focus | Impact | +|-------|--------|-------|------------|--------| +| **IC1** | Junior Designer | Components and screens | Craft skills, tools, design systems | Pixel-level quality | +| **IC2** | Product Designer | Feature areas | End-to-end design thinking, user research | Feature-level UX improvement | +| **IC3** | Senior Designer | Product area | Systems thinking, design strategy, mentoring | Product area experience | +| **IC4** | Staff Designer | Cross-product | Design vision, organizational influence | Cross-product design coherence | +| **IC5** | Principal / Head of Design | Design org | Design leadership, culture, strategy | Org-wide design quality | + +## Sales + +| Level | Titles | Scope | Key metrics | Impact | +|-------|--------|-------|------------|--------| +| **IC1** | SDR / BDR | Pipeline generation | Activities, meetings set, pipeline created | Inbound pipeline | +| **IC2** | Account Executive | Named accounts | Quota attainment, deal size, win rate | Individual quota | +| **IC3** | Senior AE | Strategic accounts | Multi-threaded deals, strategic planning | Large / strategic deals | +| **IC4** | Enterprise AE | Largest accounts | Complex deal orchestration, C-suite selling | Enterprise revenue | +| **M1** | Sales Manager | Team of 5-8 reps | Team quota, rep development | Team performance | +| **M2** | Director of Sales | Multiple teams / region | Regional strategy, hiring, forecasting | Regional outcomes | +| **M3** | VP Sales | Sales org | Go-to-market strategy, executive leadership | Revenue outcomes | + +## Marketing + +| Level | Titles | Scope | Key metrics | Impact | +|-------|--------|-------|------------|--------| +| **IC1** | Marketing Associate | Campaign execution | Activity volume, content output | Tactical support | +| **IC2** | Marketing Manager | Channel or program | Channel metrics, campaign performance | Program outcomes | +| **IC3** | Senior Marketing Manager | Multi-channel strategy | Pipeline influence, brand metrics | Marketing function outcomes | +| **IC4** | Director of Marketing | Marketing function | Revenue attribution, brand equity | Business-level marketing impact | +| **IC5** | VP Marketing | Full marketing org | Market positioning, growth strategy | Company growth | + +## Operations + +| Level | Titles | Scope | Key focus | Impact | +|-------|--------|-------|-----------|--------| +| **IC1** | Operations Associate | Process execution | Accuracy, throughput, documentation | Process compliance | +| **IC2** | Operations Manager | Workflow or function | Process improvement, vendor management | Efficiency gains | +| **IC3** | Senior Ops Manager | Cross-functional operations | Automation, scalability, analytics | Cross-team operational improvement | +| **IC4** | Director of Operations | Operations function | Strategy, org design, transformation | Operational excellence | +| **IC5** | VP Operations / COO | Full operations | Business operations strategy | Business scalability | + +## Cross-Function Level Calibration + +To check if two roles are truly at the same level across functions, they should align on: + +1. **Scope of impact** — individual, team, cross-team, org, company +2. **Ambiguity tolerance** — tasks, projects, direction, strategy +3. **Independence** — supervised, guided, independent, directing others +4. **Influence radius** — self, team, adjacent teams, org, external +5. **Time horizon** — days/weeks, months, quarters, years diff --git a/recruiting/skills/market-intelligence/SKILL.md b/recruiting/skills/market-intelligence/SKILL.md new file mode 100644 index 0000000..8d1ced0 --- /dev/null +++ b/recruiting/skills/market-intelligence/SKILL.md @@ -0,0 +1,86 @@ +--- +name: market-intelligence +description: "Research talent market conditions, compensation benchmarks, competitor hiring activity, and talent availability for recruiting decisions. Trigger with \"what's the market like for\", \"comp benchmarks\", \"salary range for\", \"who's hiring for\", \"talent supply\", \"competitor hiring\", or when the user asks about compensation, market rates, talent scarcity, or competitive landscape for a role." +--- + +# Market Intelligence + +Talent market research for informed recruiting decisions. Provides compensation benchmarks, talent supply analysis, competitor tracking, and market condition insights. + +## How It Works + +- **Standalone**: Uses web search to pull compensation data, competitor job postings, market trends, and talent supply signals. Great for quick market checks. +- **With connectors**: Enriches with data from compensation platforms, CRM contact databases, and ATS historical data on offer acceptance rates. + +## Research Dimensions + +### 1. Compensation Benchmarks + +Sources to triangulate (via web search): +- levels.fyi (tech roles, equity-heavy) +- Glassdoor (broad coverage, self-reported) +- LinkedIn Salary Insights +- Blind (anonymous, skews tech) +- Pave, Radford, Mercer (if user has access) +- H1B salary data (public, employer-specific) +- Recent job postings with published ranges (Colorado, NYC, CA transparency laws) + +When presenting comp data: +- Always show base, bonus, equity, and total compensation separately +- Adjust for geography — specify the adjustment methodology +- Show percentile ranges (25th, 50th, 75th, 90th), not just averages +- Note the data freshness and source quality +- Flag if the market is moving fast (hot roles where data lags reality) + +### 2. Talent Supply Analysis + +Assess availability by examining: +- Number of professionals with this title on LinkedIn +- Job posting volume vs. candidate volume +- Time-to-fill benchmarks for this role type +- Remote vs. in-office supply differences +- Adjacent talent pools (who could do this job with some ramp) + +Supply ratings: +| Rating | Signal | Implication | +|--------|--------|------------| +| **Abundant** | Many qualified candidates, low competition | Standard sourcing, competitive offers | +| **Balanced** | Moderate supply, moderate competition | Active sourcing needed, market-rate comp | +| **Scarce** | Few candidates, high competition | Aggressive sourcing, above-market comp, creative approaches | +| **Critical shortage** | Near-zero available talent | Consider adjacent profiles, upskilling, contract-to-hire, acqui-hire | + +### 3. Competitor Intelligence + +Track what competitors are doing: +- Open roles (what they're hiring for and at what levels) +- Compensation signals (posted ranges, Glassdoor data, Blind reports) +- Employer brand moves (blog posts, conference talks, social media) +- Layoffs or freezes (talent becoming available) +- Office/remote policies (potential lever if competitors are mandating RTO) + +### 4. Market Trends + +- Role evolution (how this role is changing, new skills emerging) +- Emerging titles and functions +- Hot skills commanding premiums +- Industry shifts affecting talent flow +- Geographic talent migration patterns + +## Output Format + +When researching market conditions, produce: + +1. **Market summary** — overall characterization of the market for this role +2. **Compensation benchmarks** — table with percentiles, adjusted for location +3. **Talent supply assessment** — availability rating with evidence +4. **Competitive landscape** — what competitors are doing in this space +5. **Recommendations** — how to position the role given market conditions + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| Data enrichment | Company employee counts, growth rates, funding data | +| ATS | Historical offer data, acceptance rates, time-to-fill by role | +| HRIS | Internal compensation data for equity analysis | +| Knowledge base | Past market research, compensation philosophy docs | diff --git a/recruiting/skills/pipeline-analytics/SKILL.md b/recruiting/skills/pipeline-analytics/SKILL.md new file mode 100644 index 0000000..e21dc0c --- /dev/null +++ b/recruiting/skills/pipeline-analytics/SKILL.md @@ -0,0 +1,90 @@ +--- +name: pipeline-analytics +description: "Analyze recruiting funnel metrics, conversion rates, time-to-fill, and pipeline velocity. Identifies bottlenecks and models scenarios to hit hiring targets. Trigger with \"pipeline metrics\", \"conversion rates\", \"time to fill\", \"funnel analysis\", \"how's the pipeline\", \"hiring velocity\", or when the user shares recruiting data, asks about hiring progress, or discusses pipeline health." +--- + +# Pipeline Analytics + +Recruiting funnel analysis and optimization. Turns raw pipeline data into actionable insights about where candidates are getting stuck, what's working, and what needs to change. + +## How It Works + +- **Standalone**: Paste pipeline data (counts by stage, time-in-stage, or just describe the situation) and get a full funnel analysis with benchmarks and recommendations. +- **With connectors**: Pull live ATS data for real-time pipeline snapshots, historical trend analysis, and automated bottleneck detection. + +## Core Metrics Framework + +### Funnel Metrics + +| Metric | Formula | Benchmark (varies by role) | What it tells you | +|--------|---------|---------------------------|-------------------| +| **Application-to-screen rate** | Screens / Applications | 15-30% | Sourcing quality and JD effectiveness | +| **Screen-to-interview rate** | Interviews / Screens | 30-50% | Recruiter calibration with hiring manager | +| **Interview-to-offer rate** | Offers / Interviews | 20-40% | Interview process effectiveness and bar clarity | +| **Offer acceptance rate** | Accepts / Offers | 70-90% | Comp competitiveness and candidate experience | +| **Overall yield** | Hires / Applications | 1-5% | End-to-end process efficiency | + +### Velocity Metrics + +| Metric | Benchmark | Red flag | +|--------|-----------|----------| +| **Time to fill** (req open → start date) | 30-60 days (varies) | > 90 days | +| **Time in stage** (per pipeline stage) | 3-7 days | > 14 days | +| **Scheduling latency** (interview requested → scheduled) | 2-3 days | > 7 days | +| **Feedback turnaround** (interview → scorecard) | 24-48 hours | > 5 days | +| **Offer turnaround** (debrief → offer extended) | 2-5 days | > 10 days | + +### Quality Metrics + +| Metric | What it measures | +|--------|-----------------| +| **New hire retention** (90-day, 1-year) | Were we hiring the right people? | +| **Hiring manager satisfaction** | Did we deliver what they needed? | +| **Candidate NPS** | How was the candidate experience? | +| **Source quality** (by channel) | Which sourcing channels produce hires? | +| **Interviewer calibration** | Do interviewer scores predict performance? | + +## Bottleneck Diagnosis + +When a stage shows poor conversion or high time-in-stage, diagnose by category: + +| Symptom | Likely causes | Recommended actions | +|---------|--------------|-------------------| +| Low screen-to-interview | Calibration gap between recruiter and HM | Joint calibration session, review recent passes and rejects | +| High time-in-stage at scheduling | Interviewer availability | Add interviewers, reduce panel size, use scheduling tools | +| Low offer acceptance | Below-market comp or slow process | Comp benchmarking, speed up offer turnaround | +| High drop-off at onsite | Poor candidate experience or bar mismatch | Candidate experience audit, interviewer training | +| Low application volume | JD issues, employer brand, or wrong channels | JD rewrite, source channel analysis | + +## Pipeline Math + +Model the pipeline needed to hit a hiring goal: + +``` +Working backward from 1 hire: + Offers needed: 1 / acceptance rate = [X] + Final interviews: offers / interview-offer = [X] + Phone screens: finals / screen-final = [X] + Applications: screens / app-screen = [X] + +At current inflow of [Y] candidates/week: + Weeks to fill: applications needed / Y = [X weeks] +``` + +## Output Format + +When analyzing a pipeline, produce: + +1. **Funnel snapshot** — current candidates by stage with conversion rates +2. **Benchmark comparison** — how each stage compares to industry benchmarks +3. **Bottleneck identification** — specific stages causing problems, with root cause hypotheses +4. **Pipeline math** — what's needed to hit the hiring goal on time +5. **Prioritized recommendations** — 3-5 actions ranked by expected impact + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| ATS | Live pipeline data, historical trends, per-role analytics | +| Calendar | Scheduling bottleneck detection, interviewer capacity analysis | +| HRIS | Headcount targets, approved roles, time-to-fill goals | diff --git a/recruiting/skills/sourcing-strategy/SKILL.md b/recruiting/skills/sourcing-strategy/SKILL.md new file mode 100644 index 0000000..9823a69 --- /dev/null +++ b/recruiting/skills/sourcing-strategy/SKILL.md @@ -0,0 +1,113 @@ +--- +name: sourcing-strategy +description: "Develop sourcing strategies to find and engage candidates, including Boolean search strings, channel recommendations, and passive candidate tactics. Trigger with \"how do I find\", \"source candidates for\", \"Boolean search\", \"where to find\", \"sourcing plan\", \"talent pool\", or when the user discusses finding candidates, building pipelines, or reaching passive talent." +--- + +# Sourcing Strategy + +Comprehensive sourcing methodology for finding and engaging candidates across all channels. Builds targeted strategies based on role characteristics, market conditions, and available resources. + +## How It Works + +- **Standalone**: Describe the role and get a complete sourcing plan with Boolean strings, channel recommendations, target companies, and outreach angles. +- **With connectors**: Scan the ATS for existing candidates, use data enrichment for contact info and company intelligence, and leverage the CRM for relationship mapping. + +## Channel Taxonomy + +### Inbound Channels + +| Channel | Best for | Expected volume | Quality signal | +|---------|----------|----------------|----------------| +| **Job boards** (Indeed, LinkedIn, etc.) | High-volume roles, active seekers | High | Lower — needs screening | +| **Career page** | Employer brand believers | Medium | Higher — intentional applicants | +| **Referrals** | All roles | Low-Medium | Highest — pre-vetted by trusted source | +| **Inbound from content** | Thought leadership roles | Low | High — engaged with your brand | + +### Outbound Channels + +| Channel | Best for | Response rate | Effort | +|---------|----------|--------------|--------| +| **LinkedIn outreach** | Most professional roles | 10-25% (if personalized) | Medium | +| **Email outreach** | When you have verified emails | 5-15% | Medium | +| **GitHub/open source** | Engineering roles | 5-10% | High (research-intensive) | +| **Conference/event networking** | Senior/niche roles | High (in-person) | High | +| **Community engagement** | Niche roles, long-term pipeline | Low (short-term) | Low (ongoing) | +| **Twitter/X** | DevRel, marketing, thought leaders | Varies | Low | + +### Third-Party Channels + +| Channel | When to use | Cost | Speed | +|---------|------------|------|-------| +| **Agency/contingency** | Urgent, senior, or niche roles | 15-25% of salary | Fast | +| **Retained search** | C-suite, confidential searches | 25-35% of salary | Medium | +| **RPO** | High-volume sustained hiring | Monthly retainer | Medium | +| **Freelance sourcers** | Overflow capacity | Per-candidate | Fast | + +## Boolean Search Construction + +### Core Pattern +``` +("[exact title]" OR "[alternate title]" OR "[related title]") +AND ("[key skill]" OR "[alternate term]") +AND ("[company type]" OR "[industry term]") +NOT ("[exclusion]" OR "[exclusion]") +``` + +### Platform-Specific Syntax + +**LinkedIn**: Use the Recruiter or Sales Navigator filters, then refine with: +``` +"senior backend engineer" AND (Python OR Go OR Rust) AND (fintech OR payments OR "financial services") +``` + +**Google X-ray** (search LinkedIn without Recruiter): +``` +site:linkedin.com/in "senior backend engineer" (Python OR Go) (fintech OR payments) -jobs -posts +``` + +**GitHub**: +``` +location:"San Francisco" language:Python followers:>50 repos:>20 +``` +Or search by contribution to specific repos relevant to the tech stack. + +**Stack Overflow**: +Search for users with high reputation in relevant tags who list a location. + +### Tips for Better Boolean Strings + +- Include synonyms and alternate titles ("Software Engineer" OR "Software Developer" OR "SDE") +- Use industry terms, not just skills ("e-commerce" vs. just "Python") +- Exclude common false positives (NOT "recruiter" NOT "sales" for engineering searches) +- Layer in company signals for quality ("Series B" OR "YC" OR specific company names) + +## Sourcing Prioritization Framework + +For any role, prioritize channels in this order: + +1. **Internal mobility** — Does anyone in the company want this role? +2. **Referrals** — Can the hiring manager's network surface candidates? +3. **ATS mining** — Silver medalists, past applicants, nurture pipeline +4. **Targeted outbound** — LinkedIn, GitHub, communities +5. **Job postings** — Optimized JD on the right boards +6. **Agency/contingency** — When speed or specialization justifies the cost + +## Output Format + +When building a sourcing strategy, produce: + +1. **Role profile** — key attributes driving the sourcing approach +2. **Channel plan** — prioritized channels with expected yield and effort +3. **Boolean strings** — 3-5 ready-to-use strings for different platforms +4. **Target companies** — tiered list with reasoning +5. **Outreach approach** — messaging angles and cadence + +## Connectors + +| Connector | Enhancement | +|-----------|------------| +| ATS | Mine existing pipeline, silver medalists, past applicants | +| Data enrichment | Contact info, company intelligence, candidate research | +| CRM | Relationship mapping, past touchpoints, nurture sequences | +| Email | Send outreach directly | +| Chat | Broadcast sourcing requests to the team for referrals |