🤖 Riksdagsmonitor — Custom Agents Guide

🎯 24 Specialised GitHub Copilot Agents Powering Democratic Transparency
🕵️ 14 Persona Agents · ⚙️ 9 Workflow-Specialist Agents · 📚 1 Shared Developer Instructions

📋 Document Owner: CEO | 🏢 Owner: Hack23 AB (Org.nr 559534-7807) | 🏷️ Classification: 🟢 Public

Overview

This repository includes custom GitHub Copilot agents specialized for different aspects of the riksdagsmonitor project. Each agent has deep expertise in its domain and can be invoked to assist with specific tasks.

Directory-level catalog: .github/agents/README.md — summarises all 24 agent files (14 persona agents + 9 workflow-specialist .agent.md + shared developer.instructions.md). This document is the long-form reference with invocation examples and per-agent skill mappings. Companion docs: SKILLS.md · .github/skills/README.md · .github/prompts/README.md · .github/workflows/README.md

Available Agents (14 Personas)

1. Security Architect (`security-architect`)

Expertise: Security architecture, ISMS compliance, threat modeling, ISO 27001/NIST CSF/CIS Controls

Use for:

Security architecture design and review
STRIDE threat modeling
Compliance framework mapping
Security control implementation
Incident response planning
Security documentation (SECURITY_ARCHITECTURE.md, THREAT_MODEL.md)

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 123,
  custom_agent: "security-architect",
  custom_instructions: "Review and update SECURITY_ARCHITECTURE.md with new controls"
})

2. Documentation Architect (`documentation-architect`)

Expertise: C4 architecture models, Mermaid diagrams, technical documentation, knowledge management

Use for:

Creating C4 architecture diagrams (Context, Container, Component)
Designing Mermaid flowcharts, sequence diagrams, state diagrams
Writing comprehensive technical documentation
Maintaining documentation portfolio (ARCHITECTURE.md, DATA_MODEL.md, etc.)
Documentation standards enforcement

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 124,
  custom_agent: "documentation-architect",
  custom_instructions: "Create comprehensive ARCHITECTURE.md with C4 models"
})

3. Quality Engineer (`quality-engineer`)

Expertise: HTML/CSS validation, accessibility testing, link checking, CI/CD quality gates

Use for:

HTML5 validation with HTMLHint
CSS3 validation and optimization
Link integrity checking with linkinator
WCAG 2.1 AA accessibility compliance
Quality workflow configuration
Performance optimization

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 125,
  custom_agent: "quality-engineer",
  custom_instructions: "Fix all HTML validation errors and ensure WCAG 2.1 AA compliance"
})

4. Frontend Specialist (`frontend-specialist`)

Expertise: Static HTML/CSS, responsive design, multi-language localization, modern frontend

Use for:

Semantic HTML5 development
Responsive CSS (Grid, Flexbox)
Multi-language website implementation (14 languages)
Cyberpunk theme design system
Cross-browser compatibility
Progressive enhancement

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 126,
  custom_agent: "frontend-specialist",
  custom_instructions: "Implement responsive navigation with support for all 14 languages"
})

5. ISMS Compliance Manager (`isms-compliance-manager`)

Expertise: ISMS policy enforcement, ISO 27001, NIST CSF 2.0, CIS Controls, audit preparation

Use for:

Compliance verification
Gap analysis
Documentation completeness checking
Audit evidence collection
Control effectiveness assessment
Compliance reporting

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 127,
  custom_agent: "isms-compliance-manager",
  custom_instructions: "Perform compliance gap analysis and verify all required documentation exists"
})

6. Deployment Specialist (`deployment-specialist`)

Expertise: GitHub Pages deployment, GitHub Actions, CI/CD security, workflow optimization

Use for:

GitHub Actions workflow design
Workflow security hardening
GitHub Pages configuration
Deployment automation
Performance optimization
Monitoring and alerting

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 128,
  custom_agent: "deployment-specialist",
  custom_instructions: "Optimize CI/CD pipeline and add security hardening"
})

7. Intelligence Operative (`intelligence-operative`)

Expertise: Political intelligence analysis, OSINT methodologies, Swedish politics, behavioral analysis

Use for:

Political data analysis and visualization dashboards
OSINT collection from Swedish government sources
Swedish political system analysis (Riksdag, 8 parties, electoral system)
Voting pattern analysis and legislative monitoring
Coalition behavior and stability assessment
Policy impact assessment
Risk assessment frameworks for democratic accountability
GDPR-compliant political data processing
Multi-language political content (14 languages)

Long-horizon expertise:

Horizon stratification (T+72h / T+7d / T+30d / T+90d / T+365d / T+1460d / election) with per-band WEP language ladder
Scenario-tree depth scaling (quarter = 4 scenarios, year = 4+5 wildcards, election-cycle = 4 scenarios × 3 coalition branches + 5 wildcards = 12 leaves)
PIR roll-forward rules for long-horizon Priority Intelligence Requirements
Cycle-rollover playbook (±30 days of election anchor) — see .github/prompts/ext/cycle-rollover.md
Runtime context via scripts/horizon-context.ts and analysis/article-types.json

Data Sources via riksdag-regering-mcp:

32 specialized tools for accessing Swedish political data
Ledamöter (MPs): Information, activities, assignments
Riksdagsdokument (Documents): Motions, questions, bills
Anföranden (Speeches): Chamber debates, statements
Voteringar (Votes): Voting records, patterns
Regeringsdokument (Government): SOU reports, propositions

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 129,
  custom_agent: "intelligence-operative",
  custom_instructions: "Create voting discipline dashboard showing party cohesion metrics for 2024"
})

8. Task Agent (`task-agent`) ✨ NEW

Expertise: Product excellence, quality assurance, ISMS compliance, Playwright testing, issue management

Use for:

Continuous product analysis (quality, UI/UX, security)
GitHub issue creation for bugs, features, improvements
HTML/CSS validation and link checking
Accessibility compliance auditing (WCAG 2.1 AA)
Security header verification
ISMS compliance tracking (ISO/NIST/CIS)
Playwright browser testing and visual regression
Agent coordination and task delegation

Key Capabilities:

Quality Assessment: HTML/CSS validation, link integrity, build health
UI/UX Evaluation: WCAG 2.1 AA audits, responsive design testing, 14-language support
Security & ISMS: ISO 27001 control verification, NIST CSF mapping, CIS Controls
Browser Testing: Playwright automation, screenshot capture, cross-browser testing
Issue Management: Well-structured GitHub issues with labels, priorities, agent assignments

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 130,
  custom_agent: "task-agent",
  custom_instructions: `
    Audit accessibility compliance for all 14 language versions.
    Create issues for any WCAG 2.1 AA violations found.
    Include Playwright screenshots as evidence.
  `
})

9. UI Enhancement Specialist (`ui-enhancement-specialist`) ✨

Expertise: Static HTML5/CSS3, responsive design, multi-language localization, WCAG 2.1 AA accessibility

Use for:

Static HTML5/CSS3 development and optimization
Cyberpunk theme design system implementation
CSS-only data visualizations (charts, heat maps, progress bars)
14-language support with RTL layouts (Arabic, Hebrew)
WCAG 2.1 AA accessibility compliance
Core Web Vitals optimization (LCP, FID, CLS)
XSS prevention and CSP enforcement
Mobile-first responsive design (320px-1440px+)

Key Capabilities:

Design System: Cyberpunk color palette, fluid typography with clamp(), CSS custom properties
Data Visualization: CSS-only progress bars, heat maps, bar charts for political metrics
Accessibility: Keyboard navigation, screen reader support, ARIA labels, color contrast 4.5:1
Multi-Language: Separate HTML files for 14 languages, hreflang SEO, RTL support
Performance: Lazy loading, CSS optimization, image optimization, Core Web Vitals
Security: No inline scripts, CSP headers, safe external links

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 131,
  custom_agent: "ui-enhancement-specialist",
  custom_instructions: `
    Implement voting discipline dashboard using CSS-only visualizations.
    Ensure WCAG 2.1 AA compliance and responsive design.
    Support all 14 languages including RTL for Arabic and Hebrew.
  `
})

10. Data Pipeline Specialist (`data-pipeline-specialist`) ✨

Expertise: CIA data consumption, ETL workflows, data validation, caching strategies, pipeline automation

Use for:

CIA export client design (19 visualization products)
JSON Schema validation against CIA schemas
Versioned caching with archival
Automated nightly data fetch workflows (02:00 CET)
Data freshness monitoring and alerting
Graceful fallback to cached data
Pipeline orchestration with GitHub Actions

Key Capabilities:

CIA Integration: Consume 19 CIA visualization product exports (election forecasts, risk assessments, OSINT)
Validation: JSON Schema validation, data quality checks, freshness monitoring
Caching: Versioned cache (current + archive), fallback strategies, data integrity
Automation: Scheduled GitHub Actions workflows, error recovery, monitoring

Addresses: Issues #18 (foundation), #19, #20 (CIA data pipeline)

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 18,
  custom_agent: "data-pipeline-specialist",
  custom_instructions: `
    Implement automated CIA data consumption pipeline:
    - Fetch 19 CIA visualization products nightly at 02:00 CET
    - Validate against CIA JSON schemas
    - Cache with version tracking and archival
    - Monitor data freshness (< 24 hours)
    - Implement graceful fallback to cached data
  `
})

11. Data Visualization Specialist (`data-visualization-specialist`) ✨

Expertise: Chart.js/D3.js, interactive dashboards, political data visualization, CIA intelligence displays

Use for:

Chart.js patterns (bar, line, scatter charts with error bars)
D3.js network diagrams (influence mapping, coalition networks)
Interactive election forecasting dashboards
Risk heat maps (45 rules × 349 MPs)
Time series visualization (50+ years historical data)
Performance optimization for large datasets
Responsive, accessible chart design

Key Capabilities:

Chart.js: Bar, line, scatter, radar charts with confidence intervals
D3.js: Force-directed graphs, network diagrams, geographic maps
Dashboards: Multi-panel CIA intelligence visualization (19 products)
Interactivity: Tooltips, zoom, pan, drill-down, real-time updates
Accessibility: WCAG 2.1 AA, keyboard navigation, screen reader support

Addresses: Issues #15, #16, #19, #20 (CIA visualization)

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 20,
  custom_agent: "data-visualization-specialist",
  custom_instructions: `
    Visualize CIA Election 2026 Forecasting Model:
    - Seat prediction bar charts with confidence intervals (±5 seats)
    - Coalition scenario probability charts (stacked bar)
    - Historical trend lines (50 years)
    - Interactive tooltips with methodology
    - Responsive design (320px-1440px+)
    - WCAG 2.1 AA accessibility
  `
})

12. Content Generator (`content-generator`) ✨ NEW

Expertise: Automated news generation, intelligence reports, multi-language content, template-based rendering

Use for:

Automated daily news articles from CIA exports
Weekly intelligence summaries
Monthly risk assessment reports
Multi-language content generation (14 languages)
Template-based markdown/HTML rendering
Scheduled content workflows
SEO optimization and front matter generation

Key Capabilities:

News Generation: Data-to-narrative transformation, inverted pyramid structure
Intelligence Reports: Executive summaries, risk assessments, coalition analyses
Multi-Language: 14-language support with RTL layouts (Arabic, Hebrew)
Automation: GitHub Actions scheduled workflows (daily 02:00 CET)
Quality: Content validation, tone consistency, factual accuracy checks

Long-horizon expertise:

Horizon stratification (T+72h → T+1460d) with per-band WEP confidence language
Registry-driven article types via scripts/horizon-context.ts and analysis/article-types.json
Cycle-rollover–aware content generation — see .github/prompts/ext/cycle-rollover.md

Addresses: Issue #17 (nightly news generation)

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 17,
  custom_agent: "content-generator",
  custom_instructions: `
    Implement nightly news generation from CIA intelligence exports:
    - Generate daily articles from 19 CIA products
    - Create election forecast updates with seat predictions
    - Generate risk assessment summaries
    - Support all 14 languages
    - Include proper YAML front matter
    - Validate content quality and factual accuracy
    - Commit to content/ directory via automated PR
  `
})

13. DevOps Engineer (`devops-engineer`) ✨ NEW

Expertise: CI/CD pipelines, GitHub Actions security, infrastructure automation, monitoring, performance optimization

Use for:

GitHub Actions workflow design and optimization
Workflow security hardening (step-security/harden-runner, SHA pinning)
GitHub Pages configuration and optimization
Automated data pipeline workflows
Performance monitoring (Core Web Vitals, Lighthouse CI)
Uptime monitoring and alerting
Build optimization (< 2 min target)
Secret management and environment configuration

Key Capabilities:

CI/CD: Multi-stage pipelines, parallel jobs, caching, artifact management
Security: Action SHA pinning, least privilege, egress auditing, secret scanning
Automation: Scheduled workflows (cron), event-driven triggers, cross-repo coordination
Monitoring: Uptime checks, build success rates, performance tracking, alerting
Optimization: Build time reduction, asset optimization, CDN configuration

Addresses: Infrastructure automation for all agents

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 200,
  custom_agent: "devops-engineer",
  custom_instructions: `
    Optimize CI/CD pipeline for riksdagsmonitor:
    - Reduce build time from 5 min to < 2 min
    - Implement GitHub Actions caching (dependencies, build)
    - Add Lighthouse CI for performance monitoring
    - Harden all workflows (SHA pinning, step-security/harden-runner)
    - Add uptime monitoring for all 14 language versions
    - Implement automated data pipeline (nightly 02:00 CET)
  `
})

14. News Journalist (`news-journalist`) ✨ NEW

Expertise: Political journalism, editorial standards, OSINT/INTOP data-driven AI-generated political intelligence, Swedish and global politics

Use for:

Political news article generation (Riksdagen, Regeringen, Myndigheter)
Editorial quality control (OSINT/INTOP-driven editorial standards)
Multi-language news coverage (14 languages)
Fact-checking and source verification
Intelligence-driven narrative reporting
SEO-optimized Schema.org structured articles

Key Capabilities:

News Generation: Data-driven articles from Riksdag/Regering MCP data
Editorial Standards: Inverted pyramid structure, balanced reporting, attribution
Multi-Language: Full translation workflow for 14 languages with RTL support
Quality: Schema.org NewsArticle, FAQPage, speakable structured data
SEO: Meta tags, Open Graph, Twitter Cards, hreflang tags

Long-horizon expertise:

Horizon stratification (T+72h → T+1460d) with per-band WEP confidence ladder
Scenario-tree branching for quarter / year / election-cycle articles
PIR roll-forward integration for forward-look coverage
Runtime context via scripts/horizon-context.ts and analysis/article-types.json

Example invocation:

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 201,
  custom_agent: "news-journalist",
  custom_instructions: `
    Generate daily political news article covering:
    - Latest government propositions
    - Opposition motions
    - Committee reports
    - Support all 14 languages
    - Include Schema.org structured data
  `
})

GitHub Copilot Coding Agent Features

All agents support modern GitHub Copilot coding agent features:

Basic Assignment (Legacy)

github-update_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 123,
  assignees: ["copilot-swe-agent[bot]"]
})

Advanced Assignment with Custom Instructions

assign_copilot_to_issue({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  issue_number: 123,
  base_ref: "main",  // Optional: specify base branch
  custom_instructions: `
    - Follow Hack23 ISMS policies
    - Use security-by-design principles
    - Update all relevant documentation
    - Ensure WCAG 2.1 AA compliance
  `
})

Direct PR Creation

create_pull_request_with_copilot({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  title: "Security Enhancement",
  body: "Implement additional security controls",
  base_ref: "main",
  custom_agent: "security-architect"
})

Job Status Tracking

const status = get_copilot_job_status({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  job_id: "abc123-def456"
});

Stacked PRs Workflow

// PR 1: Foundation
const pr1 = create_pull_request_with_copilot({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  title: "Foundation: Security documentation",
  body: "Create security architecture docs",
  base_ref: "main"
});

// PR 2: Build on PR 1
const pr2 = create_pull_request_with_copilot({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  title: "Enhancement: Security controls",
  body: "Implement security controls from documentation",
  base_ref: pr1.branch,
  custom_agent: "security-architect"
});

Agent Configuration

Repository-Level MCP Configuration

Important: Repository-level agents in .github/agents/ cannot have MCP servers configured in their YAML frontmatter. MCP servers are configured at the repository level in .github/copilot-mcp.json and are automatically available to all agents.

Each agent file's YAML frontmatter contains only:

---
name: agent-name
description: Brief description of agent's expertise
tools: ["*"]
---

All 14 primary Copilot agents in this repo use tools: ["*"] and omit mcp-servers: entirely. The agent body after the frontmatter contains detailed instructions, capabilities, examples, a standardised "Available MCP Servers" reference, standardised Copilot coding-agent tool snippets (assign_copilot_to_issue, create_pull_request_with_copilot, get_copilot_job_status, including base_ref and custom_agent), and a standardised "Related Hack23 ISMS Policies" block pointing at the governance chain rooted in Information_Security_Policy.md. Each agent file MUST remain below the 16 KB GitHub Copilot agent size limit.

MCP Server Configuration (Repository-Level Only)

MCP servers are configured in .github/copilot-mcp.json:

{
  "mcpServers": {
    "github": {
      "type": "local",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github", "--toolsets", "all", "--tools", "*"],
      "env": {
        "GITHUB_TOKEN": "${{ secrets.COPILOT_MCP_GITHUB_PERSONAL_ACCESS_TOKEN }}",
        "GITHUB_PERSONAL_ACCESS_TOKEN": "${{ secrets.COPILOT_MCP_GITHUB_PERSONAL_ACCESS_TOKEN }}",
        "GITHUB_OWNER": "Hack23",
        "GITHUB_API_URL": "https://api.githubcopilot.com/mcp/insiders"
      },
      "tools": ["*"]
    },
    "filesystem": {
      "type": "local",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/runner/work/riksdagsmonitor/riksdagsmonitor"],
      "tools": ["*"]
    },
    "git": {
      "type": "local",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-git", "--repository", "/home/runner/work/riksdagsmonitor/riksdagsmonitor"],
      "tools": ["*"]
    }
  }
}

Notes:

This configuration provides all agents with access to MCP servers
Filesystem and git paths are environment-specific (GitHub Actions runner layout)
⚠️ Security: MCP packages are not version-pinned. Consider pinning to specific versions for supply chain security

Skills Mapping by Agent

All agents automatically load relevant skills from .github/skills/ based on their expertise. Here's the recommended skill mapping for each agent:

Security Architect

Primary Skills (Core expertise):

hack23-isms-compliance - ISMS framework enforcement
security-by-design - Security architecture principles
static-site-security - Static website security
threat-modeling - STRIDE threat analysis
iso-27001-controls - ISO 27001:2022 controls
nist-csf-mapping - NIST CSF 2.0 functions
cis-controls - CIS Controls v8.1
secure-code-review - Security code review
security-documentation - ISMS documentation

Supporting Skills:

ci-cd-security - GitHub Actions security
secrets-management - Secrets and PAT management

Documentation Architect

Primary Skills:

documentation-standards - Documentation guidelines
c4-architecture-documentation - C4 model and Mermaid diagrams
security-documentation - ISMS documentation standards

Supporting Skills:

hack23-isms-compliance - Compliance requirements
threat-modeling - Threat model documentation

Quality Engineer

Primary Skills:

code-quality-checks - HTML/CSS validation, link checking
html-accessibility - WCAG 2.1 AA compliance
playwright-testing - Browser automation and testing
multi-language-localization - i18n/l10n testing

Supporting Skills:

responsive-design - Cross-device testing
static-site-security - Security validation
issue-management - Bug reporting and tracking

Frontend Specialist

Primary Skills:

responsive-design - Mobile-first responsive design
html-accessibility - WCAG 2.1 AA accessibility
multi-language-localization - 14-language support
design-system-management - Cyberpunk theme design system

Supporting Skills:

static-site-security - XSS prevention, CSP
code-quality-checks - HTML/CSS validation
political-data-visualization - CSS-only visualizations

ISMS Compliance Manager

Primary Skills:

hack23-isms-compliance - ISMS framework
iso-27001-controls - ISO 27001:2022 Annex A
nist-csf-mapping - NIST CSF 2.0
cis-controls - CIS Controls v8.1
security-documentation - Documentation standards

Supporting Skills:

security-by-design - Security principles
threat-modeling - Risk assessment
gdpr-compliance - Data protection

Deployment Specialist

Primary Skills:

github-actions-workflows - CI/CD workflow patterns
ci-cd-security - Workflow security hardening
secrets-management - GitHub secrets management

Supporting Skills:

static-site-security - Deployment security
code-quality-checks - Quality gates
hack23-isms-compliance - Compliance requirements

Intelligence Operative

Primary Skills:

riksdag-regering-mcp - 32 political data tools
political-science-analysis - Political frameworks
osint-methodologies - OSINT collection
intelligence-analysis-techniques - ACH, SWOT, Red Team
swedish-political-system - Riksdag structure
electoral-analysis - Election forecasting
behavioral-analysis - Political psychology
legislative-monitoring - Voting patterns
strategic-communication-analysis - Narrative analysis
risk-assessment-frameworks - Risk indicators
data-science-for-intelligence - Statistical analysis

Supporting Skills:

gdpr-compliance - Political data GDPR
political-data-visualization - CSS-only dashboards
multi-language-localization - 14-language content

Task Agent

Primary Skills:

issue-management - GitHub issue creation
playwright-testing - Browser automation and testing
code-quality-checks - HTML/CSS/link validation
html-accessibility - WCAG 2.1 AA audits
hack23-isms-compliance - ISMS verification

Supporting Skills:

All skills (task agent coordinates across domains)

UI Enhancement Specialist

Primary Skills:

responsive-design - Mobile-first design
design-system-management - Cyberpunk theme
political-data-visualization - CSS-only charts
html-accessibility - WCAG 2.1 AA
multi-language-localization - 14-language support

Supporting Skills:

static-site-security - XSS, CSP enforcement
code-quality-checks - Validation and testing
playwright-testing - Visual regression

Data Pipeline Specialist ✨ NEW

Primary Skills:

cia-data-integration - CIA export consumption, JSON Schema validation
data-pipeline-engineering - ETL workflows, automated scheduling
api-integration - REST/GraphQL clients, rate limiting
github-actions-workflows - Automated pipeline workflows

Supporting Skills:

performance-optimization - Data processing optimization
code-quality-checks - Data validation checks
secrets-management - API keys and credentials

Data Visualization Specialist ✨ NEW

Primary Skills:

advanced-data-visualization - Chart.js/D3.js, interactive dashboards
political-data-visualization - Political metrics visualization
responsive-design - Responsive chart layouts
html-accessibility - WCAG 2.1 AA chart accessibility
performance-optimization - Large dataset optimization

Supporting Skills:

cia-data-integration - CIA data consumption
multi-language-localization - 14-language chart labels
code-quality-checks - Chart validation

Content Generator ✨ NEW

Primary Skills:

automated-content-generation - Template-based content generation
multi-language-localization - 14-language content
html-accessibility - Accessible article structure
github-actions-workflows - Scheduled content generation

Supporting Skills:

cia-data-integration - CIA data consumption for articles
political-science-analysis - Content validation
responsive-design - Article layout
static-site-security - Content security

DevOps Engineer ✨ NEW

Primary Skills:

github-actions-workflows - CI/CD pipeline design
ci-cd-security - Workflow security hardening
secrets-management - GitHub secrets, PATs
performance-optimization - Build and runtime optimization

Supporting Skills:

static-site-security - Infrastructure security
code-quality-checks - Automated quality gates
hack23-isms-compliance - Compliance requirements
threat-modeling - Security architecture

News Journalist ✨ NEW

Primary Skills:

editorial-standards - Journalism quality standards
automated-content-generation - Template-based article generation
multi-language-localization - 14-language content
seo-optimization - Schema.org, meta tags, structured data
riksdag-regering-mcp - Swedish political data tools

Supporting Skills:

political-science-analysis - Political context
html-accessibility - Accessible article structure
responsive-design - Article layout
language-expertise - Translation quality

Best Practices

1. Choose the Right Agent

Select the agent that best matches your task:

Security tasks → Security Architect
Documentation → Documentation Architect
Quality/Testing → Quality Engineer
UI/Frontend → Frontend Specialist
Compliance → ISMS Compliance Manager
CI/CD → Deployment Specialist
Political news → News Journalist

2. Provide Clear Instructions

Use custom_instructions to give specific guidance:

custom_instructions: `
  - Follow existing patterns in src/
  - Include unit tests with 80%+ coverage
  - Update relevant documentation
  - Ensure security best practices
`

3. Leverage Skills

Agents automatically load relevant skills from .github/skills/ (91 total skills across 12 categories):

Core Infrastructure (9):

hack23-isms-compliance, security-by-design, static-site-security
ci-cd-security, documentation-standards, html-accessibility
multi-language-localization, hack23-future-architecture-standards, documentation-portfolio

Political Intelligence (11):

political-science-analysis, osint-methodologies, intelligence-analysis-techniques
swedish-political-system, electoral-analysis, behavioral-analysis
strategic-communication-analysis, legislative-monitoring
risk-assessment-frameworks, data-science-for-intelligence, gdpr-compliance

ISMS & Security (12):

cis-controls, iso-27001-controls, nist-csf-mapping
threat-modeling, secure-code-review, security-documentation
incident-response, input-validation, vulnerability-management
data-protection, ai-governance, information-security-strategy

Development & Operations (16):

c4-architecture-documentation, github-actions-workflows, code-quality-checks, secrets-management
data-pipeline-engineering, automated-content-generation, performance-optimization, api-integration
github-agentic-workflows, code-review-practices, testing-strategy, contribution-guidelines
change-management, open-source-governance, copilot-agent-patterns, product-management-patterns

UI/UX & Design (7):

responsive-design, design-system-management, political-data-visualization
advanced-data-visualization, ui-ux-design, seo-optimization, data-visualization-principles

Testing & Quality Assurance (2):

playwright-testing, issue-management

Data Integration (4):

riksdag-regering-mcp, cia-data-integration, mcp-server-development, european-parliament-api

Business & Marketing (2):

marketing, business-development

Language & Localization (1):

language-expertise

GitHub Agentic Workflows (12):

gh-aw-authentication-credentials, gh-aw-containerization, gh-aw-continuous-ai-patterns
gh-aw-firewall, gh-aw-github-actions-integration, gh-aw-logging-monitoring
gh-aw-mcp-configuration, gh-aw-mcp-gateway, gh-aw-safe-outputs
gh-aw-security-architecture, gh-aw-tools-ecosystem, gh-aw-workflow-authoring

Journalism & Media (4):

editorial-standards, investigative-journalism, prospective-news-coverage, comparative-politics-reporting

Government & Regulatory (7):

global-government-analysis, myndigheter-monitoring, regulatory-affairs
economic-policy-analysis, secure-development-lifecycle, secure-development-policy, compliance-checklist

See SKILLS.md for comprehensive skill documentation and Skills Mapping above for agent-specific recommendations.

4. Use Feature Branches

Specify base_ref for feature branch workflows:

base_ref: "feature/security-enhancements"

5. Monitor Progress

Track job status for long-running tasks:

get_copilot_job_status({
  owner: "Hack23",
  repo: "riksdagsmonitor",
  job_id: "job-id-from-assignment"
})

Agent Standards

All agents follow these standards:

Security

✅ Follow Hack23 Secure Development Policy
✅ Implement security-by-design principles
✅ Update security documentation when making changes
✅ Never introduce security vulnerabilities
✅ Use least privilege access controls

Documentation

✅ Update relevant documentation
✅ Use C4 model for architecture diagrams
✅ Create Mermaid diagrams for complex concepts
✅ Follow Hack23 documentation standards
✅ Include document control metadata

Quality

✅ Validate HTML with HTMLHint
✅ Check links with linkinator
✅ Ensure WCAG 2.1 AA accessibility
✅ Test responsive design
✅ Verify cross-browser compatibility

Compliance

✅ Map to ISO 27001/NIST CSF/CIS Controls
✅ Ensure all required docs exist
✅ Follow ISMS policies
✅ Maintain audit evidence
✅ Document compliance gaps

Troubleshooting

Agent Not Available

Verify agent file exists in .github/agents/
Check YAML frontmatter is valid
Ensure agent name matches file name

Task Not Completing

Check job status with get_copilot_job_status
Review workflow logs
Verify custom instructions are clear
Check if blocked by required reviews

Unexpected Results

Refine custom_instructions
Specify base_ref if working on feature branch
Try different agent if task doesn't match expertise
Provide more context in issue description

🌐 IMF as Primary Economic-Data Canon (Cross-Agent Rule)

Effective: 2026-04-24 · Owner: CEO · Authority: .github/aw/ECONOMIC_DATA_CONTRACT.md v2.1

Every persona agent that produces, validates, or visualises economic data MUST default to IMF as the primary economic-data source. This rule supersedes any earlier WB-centric guidance in individual agent files.

Authoritative IMF hub for agents

analysis/imf/README.md · analysis/imf/agentic-integration.md · analysis/imf/indicators-inventory.json · analysis/imf/data-dictionary.md · .github/aw/ECONOMIC_DATA_CONTRACT.md

Agent-by-agent IMF responsibility

Agent	IMF responsibility	Primary IMF dataflow
`intelligence-operative`	Cite IMF projections in scenario analysis (T+5); cross-validate IMF SWE vs SCB	WEO + FM + GFS_COFOG
`news-journalist`	Lead every economic claim with IMF citation; emit `economicProvenance.provider: imf`	WEO + FM + IFS + DOTS
`content-generator`	Render IMF dashboards in automated daily articles; vintage-aware	WEO + FM + IFS
`data-pipeline-specialist`	Own `scripts/imf-context.ts`, `imf-fetch.ts`, `imf-client.ts`, `imf-codes.ts`; vintage cache	All IMF dataflows
`data-visualization-specialist`	IMF dataflow registry as single source of truth for chart axes; vintage badge (yellow >3mo, red >6mo)	WEO + FM + ER + PCPS
`security-architect`	IMF egress allow-list (`www.imf.org`, `sdmxcentral.imf.org`); SHA-256 payload pin; vintage discipline	(security control surface)
`isms-compliance-manager`	IMF data classification: PUBLIC / no PII / GDPR DPIA short-circuit	(compliance evidence)
`documentation-architect`	IMF as first-class C4 container in ARCHITECTURE.md; mirror IMF sequence diagram from `agentic-integration.md`	(architectural surface)
`quality-engineer`	`tests/imf-context.test.ts`, `tests/imf-inventory.test.ts`, `tests/imf-precedence-contract.test.ts`	(test surface)
`devops-engineer`	IMF CLI smoke check in `mcp-setup.sh`; IMF host in `copilot-setup-steps.yml` allow-list	(operational surface)
`deployment-specialist`	IMF egress allow-list in CDN / WAF config	(deployment surface)
`frontend-specialist`	IMF data renders in dashboard HTML/CSS with vintage badge	(UI surface)
`ui-enhancement-specialist`	CSS-only IMF data visualisations (forecast cones for WEO projections)	(UI surface)
`task-agent`	IMF coverage in product analysis matrix; create issues for IMF gaps	(PM surface)

Canonical rule. Every economic claim in a Riksdagsmonitor article cites an IMF dataflow first; World Bank citations are reserved for governance, environment and social residue (the classes IMF does not publish). SCB is the Swedish-specific ground truth layer. See ECONOMIC_DATA_CONTRACT.md v2.1 for the banned-phrase list and vintage discipline (>6 mo → annotation).

🔗 Hack23 Ecosystem

🌐 Platforms	📦 Open-Source Projects	🛡️ Governance & Standards
🗳️ Riksdagsmonitor — Swedish Parliament intelligence 🇪🇺 EU Parliament Monitor — European coverage 🕵️ Citizen Intelligence Agency — political-data engine 🌐 Hack23 AB — corporate site 📰 Hack23 Blog — engineering & policy 💼 Hack23 on LinkedIn	🗳️ Hack23/riksdagsmonitor 🕵️ Hack23/cia 🇪🇺 Hack23/euparliamentmonitor 🔌 Hack23/european-parliament-mcp ✅ Hack23/cia-compliance-manager 🥋 Hack23/black-trigram 🏠 Hack23/homepage	🛡️ Hack23 ISMS-PUBLIC — public ISMS 🔒 Information Security Policy 🤖 AI Policy 🧪 Secure Development Policy 🎯 Threat Modeling Policy ⚠️ Vulnerability Management 🏷️ Classification Framework

🗳️ Empower citizens · 🔍 Strengthen democratic accountability · 🕵️ Illuminate the political process

FilesExpand file tree

AGENTS.md

Latest commit

History

AGENTS.md

File metadata and controls

🤖 Riksdagsmonitor — Custom Agents Guide

Overview

Available Agents (14 Personas)

1. Security Architect (security-architect)

2. Documentation Architect (documentation-architect)

3. Quality Engineer (quality-engineer)

4. Frontend Specialist (frontend-specialist)

5. ISMS Compliance Manager (isms-compliance-manager)

6. Deployment Specialist (deployment-specialist)

7. Intelligence Operative (intelligence-operative)

8. Task Agent (task-agent) ✨ NEW

9. UI Enhancement Specialist (ui-enhancement-specialist) ✨

10. Data Pipeline Specialist (data-pipeline-specialist) ✨

11. Data Visualization Specialist (data-visualization-specialist) ✨

12. Content Generator (content-generator) ✨ NEW

13. DevOps Engineer (devops-engineer) ✨ NEW

14. News Journalist (news-journalist) ✨ NEW

GitHub Copilot Coding Agent Features

Basic Assignment (Legacy)

Advanced Assignment with Custom Instructions

Direct PR Creation

Job Status Tracking

Stacked PRs Workflow

Agent Configuration

Repository-Level MCP Configuration

MCP Server Configuration (Repository-Level Only)

Skills Mapping by Agent

Security Architect

Documentation Architect

Quality Engineer

Frontend Specialist

ISMS Compliance Manager

Deployment Specialist

Intelligence Operative

Task Agent

UI Enhancement Specialist

Data Pipeline Specialist ✨ NEW

Data Visualization Specialist ✨ NEW

Content Generator ✨ NEW

DevOps Engineer ✨ NEW

News Journalist ✨ NEW

Best Practices

1. Choose the Right Agent

2. Provide Clear Instructions

3. Leverage Skills

4. Use Feature Branches

5. Monitor Progress

Agent Standards

Security

Documentation

Quality

Compliance

Troubleshooting

Agent Not Available

Task Not Completing

Unexpected Results

Related Documentation

🌐 IMF as Primary Economic-Data Canon (Cross-Agent Rule)

Authoritative IMF hub for agents

Agent-by-agent IMF responsibility

🔗 Hack23 Ecosystem

1. Security Architect (`security-architect`)

2. Documentation Architect (`documentation-architect`)

3. Quality Engineer (`quality-engineer`)

4. Frontend Specialist (`frontend-specialist`)

5. ISMS Compliance Manager (`isms-compliance-manager`)

6. Deployment Specialist (`deployment-specialist`)

7. Intelligence Operative (`intelligence-operative`)

8. Task Agent (`task-agent`) ✨ NEW

9. UI Enhancement Specialist (`ui-enhancement-specialist`) ✨

10. Data Pipeline Specialist (`data-pipeline-specialist`) ✨

11. Data Visualization Specialist (`data-visualization-specialist`) ✨

12. Content Generator (`content-generator`) ✨ NEW

13. DevOps Engineer (`devops-engineer`) ✨ NEW

14. News Journalist (`news-journalist`) ✨ NEW