diff --git a/apps/docs/app/global.css b/apps/docs/app/global.css
index dbcc721..c94efc0 100644
--- a/apps/docs/app/global.css
+++ b/apps/docs/app/global.css
@@ -1,3 +1,143 @@
@import "tailwindcss";
@import "fumadocs-ui/css/neutral.css";
@import "fumadocs-ui/css/preset.css";
+
+/* GitMem brand: emerald/teal accent */
+@theme {
+ --color-fd-primary: hsl(158, 64%, 32%);
+ --color-fd-primary-foreground: hsl(0, 0%, 98%);
+ --color-fd-ring: hsl(158, 64%, 42%);
+}
+
+.dark {
+ --color-fd-primary: hsl(158, 64%, 62%);
+ --color-fd-primary-foreground: hsl(0, 0%, 9%);
+ --color-fd-ring: hsl(158, 64%, 62%);
+}
+
+/* Subtle gradient in dark mode like ocean preset */
+.dark body {
+ background-image: linear-gradient(
+ rgba(16, 185, 129, 0.08),
+ transparent 20rem,
+ transparent
+ );
+ background-repeat: no-repeat;
+}
+
+/* Hero section styles for docs landing */
+.gitmem-hero {
+ text-align: center;
+ padding: 2rem 0 3rem;
+}
+
+.gitmem-hero h1 {
+ font-size: 2.5rem;
+ font-weight: 800;
+ letter-spacing: -0.025em;
+ margin-bottom: 0.75rem;
+}
+
+.gitmem-hero p {
+ font-size: 1.125rem;
+ max-width: 36rem;
+ margin: 0 auto;
+ opacity: 0.8;
+}
+
+.gitmem-hero .hero-cta {
+ display: flex;
+ gap: 0.75rem;
+ justify-content: center;
+ margin-top: 1.5rem;
+ flex-wrap: wrap;
+}
+
+.gitmem-hero .hero-cta a {
+ display: inline-flex;
+ align-items: center;
+ padding: 0.5rem 1.25rem;
+ border-radius: 0.5rem;
+ font-weight: 600;
+ font-size: 0.875rem;
+ transition: all 0.15s;
+ text-decoration: none;
+}
+
+.gitmem-hero .hero-cta a.primary {
+ background: hsl(158, 64%, 32%);
+ color: white;
+}
+
+.gitmem-hero .hero-cta a.primary:hover {
+ background: hsl(158, 64%, 28%);
+}
+
+.dark .gitmem-hero .hero-cta a.primary {
+ background: hsl(158, 64%, 52%);
+ color: hsl(0, 0%, 9%);
+}
+
+.dark .gitmem-hero .hero-cta a.primary:hover {
+ background: hsl(158, 64%, 58%);
+}
+
+.gitmem-hero .hero-cta a.secondary {
+ border: 1px solid hsl(0, 0%, 80%);
+ color: inherit;
+}
+
+.gitmem-hero .hero-cta a.secondary:hover {
+ background: hsl(0, 0%, 96%);
+}
+
+.dark .gitmem-hero .hero-cta a.secondary {
+ border-color: hsl(0, 0%, 25%);
+}
+
+.dark .gitmem-hero .hero-cta a.secondary:hover {
+ background: hsl(0, 0%, 15%);
+}
+
+/* Memory loop diagram */
+.memory-loop {
+ display: flex;
+ align-items: center;
+ justify-content: center;
+ gap: 0.25rem;
+ margin: 1.5rem auto;
+ font-size: 0.875rem;
+ font-weight: 600;
+ flex-wrap: wrap;
+}
+
+.memory-loop .step {
+ padding: 0.375rem 0.75rem;
+ border-radius: 0.375rem;
+ background: hsl(158, 64%, 92%);
+ color: hsl(158, 64%, 22%);
+}
+
+.dark .memory-loop .step {
+ background: hsla(158, 64%, 30%, 0.3);
+ color: hsl(158, 64%, 72%);
+}
+
+.memory-loop .arrow {
+ color: hsl(0, 0%, 60%);
+ font-weight: 400;
+}
+
+/* Quick install block emphasis */
+.install-block {
+ margin: 1.5rem 0;
+ padding: 1.5rem;
+ border-radius: 0.75rem;
+ border: 1px solid hsl(158, 64%, 80%);
+ background: hsl(158, 64%, 97%);
+}
+
+.dark .install-block {
+ border-color: hsla(158, 64%, 40%, 0.3);
+ background: hsla(158, 64%, 15%, 0.15);
+}
diff --git a/apps/docs/content/docs/getting-started/installation.mdx b/apps/docs/content/docs/getting-started/installation.mdx
index f2dd2a7..a4cb555 100644
--- a/apps/docs/content/docs/getting-started/installation.mdx
+++ b/apps/docs/content/docs/getting-started/installation.mdx
@@ -1,25 +1,54 @@
---
title: Installation
-description: Install GitMem via npx, global install, or MCP configuration.
+description: Install GitMem via the init wizard, manual MCP config, or global install.
---
+import { Tabs, Tab } from 'fumadocs-ui/components/tabs'
+import { Steps, Step } from 'fumadocs-ui/components/steps'
+import { Callout } from 'fumadocs-ui/components/callout'
+import { Files, File, Folder } from 'fumadocs-ui/components/files'
+import { Accordions, Accordion } from 'fumadocs-ui/components/accordion'
+
# Installation
-## Option 1: Init Wizard (Recommended)
+
+
+
+## Init Wizard
+
+The recommended way to install GitMem. One command, interactive setup, safe to re-run.
```bash
cd your-project
npx gitmem init
```
-The interactive wizard walks you through 6 steps:
-
-1. **Memory Store** — Creates `.gitmem/` with 12 starter scars
-2. **MCP Server** — Adds gitmem to `.mcp.json`
-3. **Project Instructions** — Appends memory protocols to `CLAUDE.md`
-4. **Tool Permissions** — Adds `mcp__gitmem__*` to `.claude/settings.json`
-5. **Lifecycle Hooks** — Merges session/recall hooks (preserves your existing hooks)
-6. **Gitignore** — Adds `.gitmem/` to `.gitignore`
+
+
+### Memory Store
+Creates `.gitmem/` directory with 12 starter scars covering common mistakes.
+
+
+### MCP Server
+Adds gitmem to `.mcp.json` so your AI client can discover it.
+
+
+### Project Instructions
+Appends memory protocols to `CLAUDE.md` (or your client's instruction file).
+
+
+### Tool Permissions
+Adds `mcp__gitmem__*` to `.claude/settings.json` so tools work without manual approval.
+
+
+### Lifecycle Hooks
+Merges session start and recall hooks. Preserves your existing hooks.
+
+
+### Gitignore
+Adds `.gitmem/` to `.gitignore` — memory data stays local.
+
+
### Flags
@@ -34,9 +63,28 @@ npx gitmem init --dry-run
npx gitmem init --project my-app
```
-### Idempotent
-
+
Re-running `npx gitmem init` is safe — completed steps are skipped with "Already configured."
+
+
+### What Gets Created
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
### Uninstall
@@ -51,17 +99,22 @@ Cleanly reverses everything init did. Your `.gitmem/` data is preserved by defau
npx gitmem uninstall --all
```
-## Option 2: MCP Server Only
+
+
+
+## Manual MCP Configuration
-If you want to configure the MCP server manually without the init wizard:
+If you want to configure the MCP server without the init wizard.
-### Claude Code
+
+
```bash
claude mcp add gitmem -- npx -y gitmem-mcp
```
-### Claude Desktop
+
+
Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
@@ -76,9 +129,12 @@ Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
}
```
-Restart Claude Desktop after saving.
+
+Restart Claude Desktop after saving the config file.
+
-### Cursor
+
+
Add to `.cursor/mcp.json` in your project:
@@ -93,7 +149,13 @@ Add to `.cursor/mcp.json` in your project:
}
```
-## Option 3: Global Install
+
+
+
+
+
+
+## Global Install
```bash
npm install -g gitmem-mcp
@@ -111,6 +173,15 @@ Then reference the global binary in your MCP config:
}
```
+
+Global install avoids the `npx` startup delay on each session. Useful for heavy users.
+
+
+
+
+
+---
+
## Verify Installation
After setup, ask your AI agent:
@@ -119,6 +190,10 @@ After setup, ask your AI agent:
You should see a list of available commands with your current tier and tool count.
+
+If `gitmem-help` returns a command list showing **23 tools (free tier)** or **29 tools (pro tier)**, you're good to go. Head to [Your First Session](/docs/getting-started/first-session) next.
+
+
## Environment Variables
| Variable | Default | Purpose |
@@ -127,3 +202,20 @@ You should see a list of available commands with your current tier and tool coun
| `GITMEM_DEFAULT_PROJECT` | `default` | Project namespace for scoping |
See [Configuration](/docs/getting-started/configuration) for full details.
+
+## Troubleshooting
+
+
+
+Ensure Node.js 18+ is installed (`node --version`). If behind a corporate proxy, try `npm install -g gitmem-mcp` first, then use `gitmem init` without npx.
+
+
+Check that `.mcp.json` exists in your project root and contains the gitmem server entry. For Claude Desktop, verify the config path matches your OS. Restart the client after config changes.
+
+
+Run `npx gitmem init` to add tool permissions to `.claude/settings.json`, or manually add `mcp__gitmem__*` to the `allow` array.
+
+
+Starter scars use fixed creation dates. This is cosmetic — the scars function correctly regardless of displayed age.
+
+
diff --git a/apps/docs/content/docs/index.mdx b/apps/docs/content/docs/index.mdx
index 411fc0f..55f97ac 100644
--- a/apps/docs/content/docs/index.mdx
+++ b/apps/docs/content/docs/index.mdx
@@ -3,28 +3,151 @@ title: GitMem
description: Institutional memory for AI coding agents. Memory that compounds.
---
+import { Cards, Card } from 'fumadocs-ui/components/card'
+import { Callout } from 'fumadocs-ui/components/callout'
+import { Steps, Step } from 'fumadocs-ui/components/steps'
+
+
+
# GitMem
-**Institutional memory for AI coding agents.** GitMem is an MCP server that gives AI agents the ability to learn from past mistakes, remember what worked, and share knowledge across sessions.
+Institutional memory for AI coding agents.
+Every mistake learned from. Every win remembered.
-## What is GitMem?
+
+ recall
+ →
+ work
+ →
+ learn
+ →
+ close
+ →
+ recall
+
-Every time an AI agent makes a mistake, fixes a bug, or discovers a better approach, that knowledge disappears when the session ends. GitMem captures these lessons as **scars** (mistakes to avoid), **wins** (approaches that worked), and **patterns** (reusable strategies) — then surfaces them automatically before the agent takes similar actions.
+
-## Key Features
+
+
```bash
npx gitmem init
```
One command sets up everything: `.gitmem/` directory, `.mcp.json`, `CLAUDE.md`, hooks, and permissions. Already have existing config? The wizard merges without destroying anything.
-[Get Started](/docs/getting-started) | [Concepts](/docs/concepts) | [Tool Reference](/docs/tools)
+
+
+## How It Works
+
+
+
+### Recall before acting
+
+Before consequential actions, the agent checks institutional memory. Relevant scars surface automatically with severity and counter-arguments.
+
+```
+recall({ plan: "deploy to production" })
+// → "Done ≠ Deployed ≠ Verified Working" (high severity)
+// → "Database migrations must run before code deploy" (critical)
+```
+
+
+
+### Confirm and apply
+
+Each scar must be acknowledged — **APPLYING** (with evidence), **N_A** (with reasoning), or **REFUTED** (acknowledging risk). No passive dismissal.
+
+
+
+### Learn from outcomes
+
+When something breaks or works well, capture it. Scars require counter-arguments to prevent knowledge bloat.
+
+```
+create_learning({
+ learning_type: "scar",
+ title: "OAuth redirect URI must match exactly",
+ severity: "medium",
+ counter_arguments: [
+ "In development, localhost variations are usually fine",
+ "Some providers support wildcard redirects"
+ ]
+})
+```
+
+
+
+### Close and persist
+
+Session close captures reflections, records scar usage, and persists open threads for the next session.
+
+
+
+
+## Key Features
+
+| Feature | Description |
+|---------|-------------|
+| **Automatic Recall** | Before taking action, agents check institutional memory for relevant lessons |
+| **Confirm Protocol** | Scars require active acknowledgment — APPLYING, N_A, or REFUTED |
+| **Session Continuity** | Context carries across sessions via threads, decisions, and rapport notes |
+| **Multi-Agent** | Teams of agents share the same institutional memory via `prepare_context` |
+| **MCP Native** | Works with Claude Code, Claude Desktop, Cursor, and any MCP client |
+| **Free Tier** | Local `.gitmem/` storage with 23 tools, no external dependencies |
+
+
+The free tier includes session management, scar recall, learning creation, threads, and analytics — all stored locally in `.gitmem/`. No account, no API key, no cloud dependency.
+
+
+## Explore
+
+
+
+ Install GitMem and run your first session in under 5 minutes.
+
+
+ Scars, sessions, threads, learning types, and tiers explained.
+
+
+ Complete documentation for all 29 GitMem tools.
+
+
+ Closing ceremony, multi-agent workflows, and error handling.
+
+
+ Development setup, testing tiers, and code style.
+
+
+ Release history and what's new.
+
+
diff --git a/apps/docs/content/docs/tools/analyze.mdx b/apps/docs/content/docs/tools/analyze.mdx
index d6742e3..21bb41a 100644
--- a/apps/docs/content/docs/tools/analyze.mdx
+++ b/apps/docs/content/docs/tools/analyze.mdx
@@ -16,6 +16,6 @@ Session analytics and insights engine. Returns formatted markdown by default. Us
| `lens` | `"summary"` \| `"reflections"` \| `"blindspots"` | No | Analysis lens to apply (default: summary) |
| `days` | number | No | Number of days to analyze (default: 30) |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
-| `agent` | string | No | Filter by agent identity (e.g., CLI, DAC, CODA-1) |
+| `agent` | string | No | Filter by agent identity (e.g., cli, desktop, autonomous) |
| `format` | `"text"` \| `"json"` | No | Output format: text (default, compact markdown) or json (raw data) |
diff --git a/apps/docs/content/docs/tools/confirm-scars.mdx b/apps/docs/content/docs/tools/confirm-scars.mdx
index 3efe88b..5ef8604 100644
--- a/apps/docs/content/docs/tools/confirm-scars.mdx
+++ b/apps/docs/content/docs/tools/confirm-scars.mdx
@@ -22,4 +22,5 @@ Confirm surfaced scars with APPLYING/N_A/REFUTED decisions and evidence. REQUIRE
| `scar_id` | string | Yes | UUID of the surfaced scar (from recall result) |
| `decision` | `"APPLYING"` \| `"N_A"` \| `"REFUTED"` | Yes | APPLYING: scar is relevant, evidence of compliance. N_A: scar doesn't apply, explain why. REFUTED: overriding scar, acknowledge risk. |
| `evidence` | string | Yes | Past-tense evidence (APPLYING), scenario comparison (N_A), or risk acknowledgment (REFUTED). Minimum 50 characters. |
+| `relevance` | `"high"` \| `"low"` \| `"noise"` | No | How relevant was this scar to your plan? high=directly applicable, low=tangentially related, noise=not relevant to this context. Helps improve future recall quality. |
diff --git a/apps/docs/content/docs/tools/create-learning.mdx b/apps/docs/content/docs/tools/create-learning.mdx
index 779e3c0..70a911b 100644
--- a/apps/docs/content/docs/tools/create-learning.mdx
+++ b/apps/docs/content/docs/tools/create-learning.mdx
@@ -7,15 +7,15 @@ description: "Create scar, win, or pattern entry in institutional memory."
**Tier:** Free · **Aliases:** `gitmem-cl`, `gm-scar`
-Create scar, win, or pattern entry in institutional memory
+Create scar, win, or pattern entry in institutional memory. Frame as 'what we now know' — lead with the factual/architectural discovery, not what went wrong. Good: 'Fine-grained PATs are scoped to one resource owner'. Bad: 'Should have checked PAT type first'.
## Parameters
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `learning_type` | `"scar"` \| `"win"` \| `"pattern"` \| `"anti_pattern"` | Yes | Type of learning |
-| `title` | string | Yes | Learning title |
-| `description` | string | Yes | Detailed description |
+| `title` | string | Yes | Frame as a knowledge discovery — what we now know. Lead with the factual insight, not self-criticism. |
+| `description` | string | Yes | Detailed description. Include the architectural/behavioral fact that makes this retrievable by domain. |
| `severity` | `"critical"` \| `"high"` \| `"medium"` \| `"low"` | No | Severity level (required for scars) |
| `scar_type` | `"process"` \| `"incident"` \| `"context"` | No | Scar type (process, incident, or context). Defaults to 'process'. |
| `counter_arguments` | string[] | No | Counter-arguments for scars (min 2 required) |
diff --git a/apps/docs/content/docs/tools/graph-traverse.mdx b/apps/docs/content/docs/tools/graph-traverse.mdx
index 162ba93..0125f66 100644
--- a/apps/docs/content/docs/tools/graph-traverse.mdx
+++ b/apps/docs/content/docs/tools/graph-traverse.mdx
@@ -7,14 +7,14 @@ description: "Traverse the knowledge graph over institutional memory triples."
**Tier:** Pro · **Aliases:** `gitmem-graph`, `gm-graph`
-Traverse the knowledge graph over institutional memory triples. Answers: 'show me everything connected to PROJ-123', 'what did CLI produce', 'trace this decision back', 'which issues produced the most learnings'. Four lenses: connected_to, produced_by, provenance, stats.
+Traverse the knowledge graph over institutional memory triples. Answers: 'show me everything connected to this issue', 'what did this agent produce', 'trace this decision back', 'which issues produced the most learnings'. Four lenses: connected_to, produced_by, provenance, stats.
## Parameters
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `lens` | `"connected_to"` \| `"produced_by"` \| `"provenance"` \| `"stats"` | Yes | Traversal mode: connected_to (all connections to a node), produced_by (what an agent/persona produced), provenance (trace origin chain), stats (aggregate counts) |
-| `node` | string | No | Starting node. Examples: 'PROJ-123', 'CLI', 'Scar: Done ≠ Deployed'. Required for all lenses except stats. |
+| `node` | string | No | Starting node. Examples: 'PROJ-123', 'cli', 'Scar: Done ≠ Deployed'. Required for all lenses except stats. |
| `predicate` | `"created_in"` \| `"influenced_by"` \| `"supersedes"` \| `"demonstrates"` | No | Filter by predicate (optional) |
| `depth` | number | No | Max chain depth for provenance lens (default: 3) |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
diff --git a/apps/docs/content/docs/tools/recall.mdx b/apps/docs/content/docs/tools/recall.mdx
index 871e14f..5955491 100644
--- a/apps/docs/content/docs/tools/recall.mdx
+++ b/apps/docs/content/docs/tools/recall.mdx
@@ -7,7 +7,7 @@ description: "Check institutional memory for relevant scars before taking action
**Tier:** Free · **Aliases:** `gitmem-r`
-Check institutional memory for relevant scars before taking action. Returns matching scars and their lessons.
+Check institutional memory for relevant scars before taking action. Returns matching scars and their lessons. Integrates variant assignment when issue_id provided.
## Parameters
@@ -17,4 +17,5 @@ Check institutional memory for relevant scars before taking action. Returns matc
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
| `match_count` | number | No | Number of scars to return (default: 3) |
| `issue_id` | string | No | Linear issue identifier for variant assignment (e.g., 'PROJ-123'). When provided, scars with variants will be randomly assigned and formatted accordingly. |
+| `similarity_threshold` | number | No | Minimum similarity score (0-1) to include results. Weak matches below threshold are suppressed. Default: 0.4 (free tier BM25), 0.35 (pro tier embeddings). |
diff --git a/apps/docs/content/docs/tools/record-scar-usage-batch.mdx b/apps/docs/content/docs/tools/record-scar-usage-batch.mdx
index 4290447..c447a5c 100644
--- a/apps/docs/content/docs/tools/record-scar-usage-batch.mdx
+++ b/apps/docs/content/docs/tools/record-scar-usage-batch.mdx
@@ -29,5 +29,5 @@ Track multiple scar applications in a single batch operation (reduces session cl
| `reference_context` | string | Yes | How the scar was applied (1-2 sentences) |
| `execution_successful` | boolean | No | Whether the task succeeded after applying scar |
| `session_id` | string | No | GitMem session UUID (for non-issue session tracking) |
-| `agent` | string | No | Agent identity (CLI, DAC, CODA-1, etc.) |
+| `agent` | string | No | Agent identity (e.g., cli, desktop, autonomous) |
diff --git a/apps/docs/content/docs/tools/record-scar-usage.mdx b/apps/docs/content/docs/tools/record-scar-usage.mdx
index cf72943..a8d5f01 100644
--- a/apps/docs/content/docs/tools/record-scar-usage.mdx
+++ b/apps/docs/content/docs/tools/record-scar-usage.mdx
@@ -13,7 +13,7 @@ Track scar application for effectiveness measurement
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
-| `scar_id` | string | Yes | UUID of the scar |
+| `scar_id` | string | Yes | UUID of the scar |
| `issue_id` | string | No | Linear issue UUID |
| `issue_identifier` | string | No | Linear issue identifier (e.g., PROJ-123) |
| `surfaced_at` | string | Yes | ISO timestamp when scar was retrieved |
@@ -22,6 +22,6 @@ Track scar application for effectiveness measurement
| `reference_context` | string | Yes | How the scar was applied (1-2 sentences) |
| `execution_successful` | boolean | No | Whether the task succeeded after applying scar |
| `session_id` | string | No | GitMem session UUID (for non-issue session tracking) |
-| `agent` | string | No | Agent identity (CLI, DAC, CODA-1, etc.) |
+| `agent` | string | No | Agent identity (e.g., cli, desktop, autonomous) |
| `variant_id` | string | No | UUID of the assigned variant from scar_enforcement_variants (for A/B testing) |
diff --git a/apps/docs/content/docs/tools/session-start.mdx b/apps/docs/content/docs/tools/session-start.mdx
index c0485ce..21d2d04 100644
--- a/apps/docs/content/docs/tools/session-start.mdx
+++ b/apps/docs/content/docs/tools/session-start.mdx
@@ -13,7 +13,7 @@ Initialize session, detect agent, load institutional context (last session, rece
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
-| `agent_identity` | `"CLI"` \| `"DAC"` \| `"CODA-1"` \| `"Brain_Local"` \| `"Brain_Cloud"` | No | Override agent identity (auto-detects if not provided) |
+| `agent_identity` | `"cli"` \| `"desktop"` \| `"autonomous"` \| `"local"` \| `"cloud"` | No | Override agent identity (auto-detects if not provided) |
| `linear_issue` | string | No | Current Linear issue identifier (e.g., PROJ-123) |
| `issue_title` | string | No | Issue title for scar context |
| `issue_description` | string | No | Issue description for scar context |
diff --git a/apps/docs/lib/layout.shared.tsx b/apps/docs/lib/layout.shared.tsx
index 8d42244..5180f82 100644
--- a/apps/docs/lib/layout.shared.tsx
+++ b/apps/docs/lib/layout.shared.tsx
@@ -18,7 +18,7 @@ export function baseOptions(): BaseLayoutProps {
},
{
text: "GitHub",
- url: "https://github.com/nTEG-dev/gitmem",
+ url: "https://github.com/gitmem-dev/gitmem",
external: true,
},
{
diff --git a/apps/docs/public/llms-full.txt b/apps/docs/public/llms-full.txt
index 1a4f940..9b45035 100644
--- a/apps/docs/public/llms-full.txt
+++ b/apps/docs/public/llms-full.txt
@@ -1,7 +1,7 @@
# GitMem — Complete Documentation
> GitMem is an MCP server that gives AI coding agents persistent institutional memory.
-> Install: `npx gitmem init` or add MCP server `npx -y gitmem-mcp`
+> Install: `npx gitmem-mcp init` or add MCP server `npx -y gitmem-mcp`
> Source: https://gitmem.ai
> Index: https://gitmem.ai/llms.txt
@@ -25,17 +25,17 @@ Description: GitMem release history.
### Docs
-- Updated README to feature one-command setup, added gitmem.dev link
+- Updated README to feature one-command setup, added gitmem.ai link
- Updated docs site installation and getting-started pages
## v1.0.1 (2026-02-14)
### Bug Fixes
-- **OD-590:** Fixed stdout corruption — 22 `console.log` calls replaced with `console.error` in check.ts
-- **OD-554:** Error messages now surface in `create_learning` and `record_scar_usage` responses via `errors[]` array
-- **OD-548:** Session ID validation added to `session_close` — returns clear error with UUID format example instead of confusing DB lookup failure
-- **OD-640:** Removed hardcoded project enum — project names are now free-form strings
+- Fixed stdout corruption — 22 `console.log` calls replaced with `console.error` in check.ts
+- Error messages now surface in `create_learning` and `record_scar_usage` responses via `errors[]` array
+- Session ID validation added to `session_close` — returns clear error with UUID format example instead of confusing DB lookup failure
+- Removed hardcoded project enum — project names are now free-form strings
### Tests
@@ -394,7 +394,7 @@ Set via: `GITMEM_TIER=free` (default)
- **Tools:** 29 tools (adds batch, transcripts, cache management)
- **Best for:** Teams, multi-machine workflows, production use
-Pro is currently in development. [Join the mailing list](https://gitmem.dev) to get notified when it launches.
+Pro is currently in development. [Join the mailing list](https://gitmem.ai) to get notified when it launches.
## Additional Tools (Pro)
@@ -608,7 +608,7 @@ Add `.gitmem/` to your `.gitignore` — session data is machine-specific.
### Pro Tier (Coming Soon)
-Pro will add Supabase PostgreSQL with pgvector for semantic search, cross-machine sync, and shared team memory. [Join the mailing list](https://gitmem.dev) to get notified.
+Pro will add Supabase PostgreSQL with pgvector for semantic search, cross-machine sync, and shared team memory. [Join the mailing list](https://gitmem.ai) to get notified.
## Project Namespaces
@@ -746,7 +746,7 @@ Pro adds cloud-powered features for teams and power users:
- You need session transcript storage
- You're running multi-agent workflows across environments
-[Join the mailing list](https://gitmem.dev) to get notified when Pro launches.
+[Join the mailing list](https://gitmem.ai) to get notified when Pro launches.
========================================================================
# Overview
@@ -760,7 +760,7 @@ GitMem gives your AI coding agents institutional memory — the ability to learn
## Prerequisites
-- **Node.js 22+** (required by Fumadocs MDX)
+- **Node.js 18+** (for `npx gitmem init`)
- An MCP-compatible client: [Claude Code](https://claude.com/claude-code), [Claude Desktop](https://claude.ai/download), or [Cursor](https://cursor.sh)
## Quick Start (30 seconds)
@@ -796,26 +796,55 @@ Already have `.mcp.json`, `CLAUDE.md`, or hooks? The wizard merges without destr
========================================================================
# Installation
URL: https://gitmem.ai/docs/getting-started/installation
-Description: Install GitMem via npx, global install, or MCP configuration.
+Description: Install GitMem via the init wizard, manual MCP config, or global install.
========================================================================
+import { Tabs, Tab } from 'fumadocs-ui/components/tabs'
+import { Steps, Step } from 'fumadocs-ui/components/steps'
+import { Callout } from 'fumadocs-ui/components/callout'
+import { Files, File, Folder } from 'fumadocs-ui/components/files'
+import { Accordions, Accordion } from 'fumadocs-ui/components/accordion'
+
# Installation
-## Option 1: Init Wizard (Recommended)
+
+
+
+## Init Wizard
+
+The recommended way to install GitMem. One command, interactive setup, safe to re-run.
```bash
cd your-project
npx gitmem init
```
-The interactive wizard walks you through 6 steps:
-
-1. **Memory Store** — Creates `.gitmem/` with 12 starter scars
-2. **MCP Server** — Adds gitmem to `.mcp.json`
-3. **Project Instructions** — Appends memory protocols to `CLAUDE.md`
-4. **Tool Permissions** — Adds `mcp__gitmem__*` to `.claude/settings.json`
-5. **Lifecycle Hooks** — Merges session/recall hooks (preserves your existing hooks)
-6. **Gitignore** — Adds `.gitmem/` to `.gitignore`
+
+
+### Memory Store
+Creates `.gitmem/` directory with 12 starter scars covering common mistakes.
+
+
+### MCP Server
+Adds gitmem to `.mcp.json` so your AI client can discover it.
+
+
+### Project Instructions
+Appends memory protocols to `CLAUDE.md` (or your client's instruction file).
+
+
+### Tool Permissions
+Adds `mcp__gitmem__*` to `.claude/settings.json` so tools work without manual approval.
+
+
+### Lifecycle Hooks
+Merges session start and recall hooks. Preserves your existing hooks.
+
+
+### Gitignore
+Adds `.gitmem/` to `.gitignore` — memory data stays local.
+
+
### Flags
@@ -830,9 +859,28 @@ npx gitmem init --dry-run
npx gitmem init --project my-app
```
-### Idempotent
-
+
Re-running `npx gitmem init` is safe — completed steps are skipped with "Already configured."
+
+
+### What Gets Created
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
### Uninstall
@@ -847,17 +895,22 @@ Cleanly reverses everything init did. Your `.gitmem/` data is preserved by defau
npx gitmem uninstall --all
```
-## Option 2: MCP Server Only
+
+
+
+## Manual MCP Configuration
-If you want to configure the MCP server manually without the init wizard:
+If you want to configure the MCP server without the init wizard.
-### Claude Code
+
+
```bash
claude mcp add gitmem -- npx -y gitmem-mcp
```
-### Claude Desktop
+
+
Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
@@ -872,9 +925,12 @@ Add to `~/Library/Application Support/Claude/claude_desktop_config.json`:
}
```
-Restart Claude Desktop after saving.
+
+Restart Claude Desktop after saving the config file.
+
-### Cursor
+
+
Add to `.cursor/mcp.json` in your project:
@@ -889,7 +945,13 @@ Add to `.cursor/mcp.json` in your project:
}
```
-## Option 3: Global Install
+
+
+
+
+
+
+## Global Install
```bash
npm install -g gitmem-mcp
@@ -907,6 +969,15 @@ Then reference the global binary in your MCP config:
}
```
+
+Global install avoids the `npx` startup delay on each session. Useful for heavy users.
+
+
+
+
+
+---
+
## Verify Installation
After setup, ask your AI agent:
@@ -915,6 +986,10 @@ After setup, ask your AI agent:
You should see a list of available commands with your current tier and tool count.
+
+If `gitmem-help` returns a command list showing **23 tools (free tier)** or **29 tools (pro tier)**, you're good to go. Head to [Your First Session](/docs/getting-started/first-session) next.
+
+
## Environment Variables
| Variable | Default | Purpose |
@@ -924,6 +999,23 @@ You should see a list of available commands with your current tier and tool coun
See [Configuration](/docs/getting-started/configuration) for full details.
+## Troubleshooting
+
+
+
+Ensure Node.js 18+ is installed (`node --version`). If behind a corporate proxy, try `npm install -g gitmem-mcp` first, then use `gitmem init` without npx.
+
+
+Check that `.mcp.json` exists in your project root and contains the gitmem server entry. For Claude Desktop, verify the config path matches your OS. Restart the client after config changes.
+
+
+Run `npx gitmem init` to add tool permissions to `.claude/settings.json`, or manually add `mcp__gitmem__*` to the `allow` array.
+
+
+Starter scars use fixed creation dates. This is cosmetic — the scars function correctly regardless of displayed age.
+
+
+
========================================================================
# Closing Ceremony
URL: https://gitmem.ai/docs/guides/closing-ceremony
@@ -1135,31 +1227,154 @@ URL: https://gitmem.ai/docs
Description: Institutional memory for AI coding agents. Memory that compounds.
========================================================================
+import { Cards, Card } from 'fumadocs-ui/components/card'
+import { Callout } from 'fumadocs-ui/components/callout'
+import { Steps, Step } from 'fumadocs-ui/components/steps'
+
+
+
# GitMem
-**Institutional memory for AI coding agents.** GitMem is an MCP server that gives AI agents the ability to learn from past mistakes, remember what worked, and share knowledge across sessions.
+Institutional memory for AI coding agents.
+Every mistake learned from. Every win remembered.
-## What is GitMem?
+
+ recall
+ →
+ work
+ →
+ learn
+ →
+ close
+ →
+ recall
+
-Every time an AI agent makes a mistake, fixes a bug, or discovers a better approach, that knowledge disappears when the session ends. GitMem captures these lessons as **scars** (mistakes to avoid), **wins** (approaches that worked), and **patterns** (reusable strategies) — then surfaces them automatically before the agent takes similar actions.
+
-## Key Features
+
+
```bash
npx gitmem init
```
One command sets up everything: `.gitmem/` directory, `.mcp.json`, `CLAUDE.md`, hooks, and permissions. Already have existing config? The wizard merges without destroying anything.
-[Get Started](/docs/getting-started) | [Concepts](/docs/concepts) | [Tool Reference](/docs/tools)
+
+
+## How It Works
+
+
+
+### Recall before acting
+
+Before consequential actions, the agent checks institutional memory. Relevant scars surface automatically with severity and counter-arguments.
+
+```
+recall({ plan: "deploy to production" })
+// → "Done ≠ Deployed ≠ Verified Working" (high severity)
+// → "Database migrations must run before code deploy" (critical)
+```
+
+
+
+### Confirm and apply
+
+Each scar must be acknowledged — **APPLYING** (with evidence), **N_A** (with reasoning), or **REFUTED** (acknowledging risk). No passive dismissal.
+
+
+
+### Learn from outcomes
+
+When something breaks or works well, capture it. Scars require counter-arguments to prevent knowledge bloat.
+
+```
+create_learning({
+ learning_type: "scar",
+ title: "OAuth redirect URI must match exactly",
+ severity: "medium",
+ counter_arguments: [
+ "In development, localhost variations are usually fine",
+ "Some providers support wildcard redirects"
+ ]
+})
+```
+
+
+
+### Close and persist
+
+Session close captures reflections, records scar usage, and persists open threads for the next session.
+
+
+
+
+## Key Features
+
+| Feature | Description |
+|---------|-------------|
+| **Automatic Recall** | Before taking action, agents check institutional memory for relevant lessons |
+| **Confirm Protocol** | Scars require active acknowledgment — APPLYING, N_A, or REFUTED |
+| **Session Continuity** | Context carries across sessions via threads, decisions, and rapport notes |
+| **Multi-Agent** | Teams of agents share the same institutional memory via `prepare_context` |
+| **MCP Native** | Works with Claude Code, Claude Desktop, Cursor, and any MCP client |
+| **Free Tier** | Local `.gitmem/` storage with 23 tools, no external dependencies |
+
+
+The free tier includes session management, scar recall, learning creation, threads, and analytics — all stored locally in `.gitmem/`. No account, no API key, no cloud dependency.
+
+
+## Explore
+
+
+
+ Install GitMem and run your first session in under 5 minutes.
+
+
+ Scars, sessions, threads, learning types, and tiers explained.
+
+
+ Complete documentation for all 29 GitMem tools.
+
+
+ Closing ceremony, multi-agent workflows, and error handling.
+
+
+ Development setup, testing tiers, and code style.
+
+
+ Release history and what's new.
+
+
========================================================================
# absorb_observations
@@ -1208,7 +1423,7 @@ Session analytics and insights engine. Returns formatted markdown by default. Us
| `lens` | `"summary"` \| `"reflections"` \| `"blindspots"` | No | Analysis lens to apply (default: summary) |
| `days` | number | No | Number of days to analyze (default: 30) |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
-| `agent` | string | No | Filter by agent identity (e.g., CLI, DAC, CODA-1) |
+| `agent` | string | No | Filter by agent identity (e.g., cli, desktop, autonomous) |
| `format` | `"text"` \| `"json"` | No | Output format: text (default, compact markdown) or json (raw data) |
========================================================================
@@ -1228,7 +1443,7 @@ Archives a learning (scar/win/pattern) by setting is_active=false and recording
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `id` | string | Yes | UUID of the learning to archive |
-| `reason` | string | No | Optional reason for archiving (e.g., 'superseded by OD-XXX', 'no longer relevant') |
+| `reason` | string | No | Optional reason for archiving (e.g., 'superseded by PROJ-123', 'no longer relevant') |
========================================================================
# gitmem-cache-flush
@@ -1328,6 +1543,7 @@ Confirm surfaced scars with APPLYING/N_A/REFUTED decisions and evidence. REQUIRE
| `scar_id` | string | Yes | UUID of the surfaced scar (from recall result) |
| `decision` | `"APPLYING"` \| `"N_A"` \| `"REFUTED"` | Yes | APPLYING: scar is relevant, evidence of compliance. N_A: scar doesn't apply, explain why. REFUTED: overriding scar, acknowledge risk. |
| `evidence` | string | Yes | Past-tense evidence (APPLYING), scenario comparison (N_A), or risk acknowledgment (REFUTED). Minimum 50 characters. |
+| `relevance` | `"high"` \| `"low"` \| `"noise"` | No | How relevant was this scar to your plan? high=directly applicable, low=tangentially related, noise=not relevant to this context. Helps improve future recall quality. |
========================================================================
# create_decision
@@ -1365,15 +1581,15 @@ Description: Create scar, win, or pattern entry in institutional memory.
**Tier:** Free · **Aliases:** `gitmem-cl`, `gm-scar`
-Create scar, win, or pattern entry in institutional memory
+Create scar, win, or pattern entry in institutional memory. Frame as 'what we now know' — lead with the factual/architectural discovery, not what went wrong. Good: 'Fine-grained PATs are scoped to one resource owner'. Bad: 'Should have checked PAT type first'.
## Parameters
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `learning_type` | `"scar"` \| `"win"` \| `"pattern"` \| `"anti_pattern"` | Yes | Type of learning |
-| `title` | string | Yes | Learning title |
-| `description` | string | Yes | Detailed description |
+| `title` | string | Yes | Frame as a knowledge discovery — what we now know. Lead with the factual insight, not self-criticism. |
+| `description` | string | Yes | Detailed description. Include the architectural/behavioral fact that makes this retrievable by domain. |
| `severity` | `"critical"` \| `"high"` \| `"medium"` \| `"low"` | No | Severity level (required for scars) |
| `scar_type` | `"process"` \| `"incident"` \| `"context"` | No | Scar type (process, incident, or context). Defaults to 'process'. |
| `counter_arguments` | string[] | No | Counter-arguments for scars (min 2 required) |
@@ -1402,7 +1618,7 @@ Create an open thread to track unresolved work across sessions. Includes semanti
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `text` | string | Yes | Thread description — what needs to be tracked or resolved |
-| `linear_issue` | string | No | Associated Linear issue (e.g., OD-XXX) |
+| `linear_issue` | string | No | Associated Linear issue (e.g., PROJ-123) |
========================================================================
# dismiss_suggestion
@@ -1425,14 +1641,14 @@ Dismiss a suggested thread. Incremented dismiss count — suggestions dismissed
========================================================================
# get_transcript
URL: https://gitmem.ai/docs/tools/get-transcript
-Description: Retrieve a session transcript from storage (OD-467).
+Description: Retrieve a session transcript from storage.
========================================================================
# get_transcript
**Tier:** Dev · **Aliases:** `gitmem-gt`
-Retrieve a session transcript from storage (OD-467)
+Retrieve a session transcript from storage
## Parameters
@@ -1450,14 +1666,14 @@ Description: Traverse the knowledge graph over institutional memory triples.
**Tier:** Pro · **Aliases:** `gitmem-graph`, `gm-graph`
-Traverse the knowledge graph over institutional memory triples. Answers: 'show me everything connected to OD-466', 'what did CLI produce', 'trace this decision back', 'which issues produced the most learnings'. Four lenses: connected_to, produced_by, provenance, stats.
+Traverse the knowledge graph over institutional memory triples. Answers: 'show me everything connected to this issue', 'what did this agent produce', 'trace this decision back', 'which issues produced the most learnings'. Four lenses: connected_to, produced_by, provenance, stats.
## Parameters
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| `lens` | `"connected_to"` \| `"produced_by"` \| `"provenance"` \| `"stats"` | Yes | Traversal mode: connected_to (all connections to a node), produced_by (what an agent/persona produced), provenance (trace origin chain), stats (aggregate counts) |
-| `node` | string | No | Starting node. Examples: 'OD-466', 'CLI', 'Scar: Done ≠ Deployed'. Required for all lenses except stats. |
+| `node` | string | No | Starting node. Examples: 'PROJ-123', 'cli', 'Scar: Done ≠ Deployed'. Required for all lenses except stats. |
| `predicate` | `"created_in"` \| `"influenced_by"` \| `"supersedes"` \| `"demonstrates"` | No | Filter by predicate (optional) |
| `depth` | number | No | Max chain depth for provenance lens (default: 3) |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
@@ -1665,7 +1881,7 @@ Description: Check institutional memory for relevant scars before taking action.
**Tier:** Free · **Aliases:** `gitmem-r`
-Check institutional memory for relevant scars before taking action. Returns matching scars and their lessons.
+Check institutional memory for relevant scars before taking action. Returns matching scars and their lessons. Integrates variant assignment when issue_id provided.
## Parameters
@@ -1674,7 +1890,8 @@ Check institutional memory for relevant scars before taking action. Returns matc
| `plan` | string | Yes | What you're about to do (e.g., 'implement auth layer', 'deploy to production') |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
| `match_count` | number | No | Number of scars to return (default: 3) |
-| `issue_id` | string | No | Linear issue identifier for variant assignment (e.g., 'OD-525'). When provided, scars with variants will be randomly assigned and formatted accordingly. |
+| `issue_id` | string | No | Linear issue identifier for variant assignment (e.g., 'PROJ-123'). When provided, scars with variants will be randomly assigned and formatted accordingly. |
+| `similarity_threshold` | number | No | Minimum similarity score (0-1) to include results. Weak matches below threshold are suppressed. Default: 0.4 (free tier BM25), 0.35 (pro tier embeddings). |
========================================================================
# record_scar_usage_batch
@@ -1701,14 +1918,14 @@ Track multiple scar applications in a single batch operation (reduces session cl
|-------|------|----------|-------------|
| `scar_identifier` | string | Yes | UUID or title/description of scar (tool resolves to UUID) |
| `issue_id` | string | No | Linear issue UUID |
-| `issue_identifier` | string | No | Linear issue identifier (e.g., OD-XXX) |
+| `issue_identifier` | string | No | Linear issue identifier (e.g., PROJ-123) |
| `surfaced_at` | string | Yes | ISO timestamp when scar was retrieved |
| `acknowledged_at` | string | No | ISO timestamp when scar was acknowledged |
| `reference_type` | `"explicit"` \| `"implicit"` \| `"acknowledged"` \| `"refuted"` \| `"none"` | Yes | How the scar was referenced |
| `reference_context` | string | Yes | How the scar was applied (1-2 sentences) |
| `execution_successful` | boolean | No | Whether the task succeeded after applying scar |
| `session_id` | string | No | GitMem session UUID (for non-issue session tracking) |
-| `agent` | string | No | Agent identity (CLI, DAC, CODA-1, etc.) |
+| `agent` | string | No | Agent identity (e.g., cli, desktop, autonomous) |
========================================================================
# record_scar_usage
@@ -1726,16 +1943,16 @@ Track scar application for effectiveness measurement
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
-| `scar_id` | string | Yes | UUID of the scar from orchestra_learnings |
+| `scar_id` | string | Yes | UUID of the scar |
| `issue_id` | string | No | Linear issue UUID |
-| `issue_identifier` | string | No | Linear issue identifier (e.g., OD-XXX) |
+| `issue_identifier` | string | No | Linear issue identifier (e.g., PROJ-123) |
| `surfaced_at` | string | Yes | ISO timestamp when scar was retrieved |
| `acknowledged_at` | string | No | ISO timestamp when scar was acknowledged |
| `reference_type` | `"explicit"` \| `"implicit"` \| `"acknowledged"` \| `"refuted"` \| `"none"` | Yes | How the scar was referenced |
| `reference_context` | string | Yes | How the scar was applied (1-2 sentences) |
| `execution_successful` | boolean | No | Whether the task succeeded after applying scar |
| `session_id` | string | No | GitMem session UUID (for non-issue session tracking) |
-| `agent` | string | No | Agent identity (CLI, DAC, CODA-1, etc.) |
+| `agent` | string | No | Agent identity (e.g., cli, desktop, autonomous) |
| `variant_id` | string | No | UUID of the assigned variant from scar_enforcement_variants (for A/B testing) |
========================================================================
@@ -1761,14 +1978,14 @@ Mark an open thread as resolved. Use thread_id for exact match or text_match for
========================================================================
# save_transcript
URL: https://gitmem.ai/docs/tools/save-transcript
-Description: Save full session transcript to storage for training data and post-mortems (OD-467).
+Description: Save full session transcript to storage for training data and post-mortems.
========================================================================
# save_transcript
**Tier:** Dev · **Aliases:** `gitmem-st`
-Save full session transcript to storage for training data and post-mortems (OD-467)
+Save full session transcript to storage for training data and post-mortems
## Parameters
@@ -1798,7 +2015,7 @@ Semantic search over session transcript chunks. Generates embedding for query an
| `query` | string | Yes | Natural language search query (e.g., 'deployment verification discussion', 'what was decided about caching') |
| `match_count` | number | No | Maximum number of chunks to return (default: 10, max: 50) |
| `similarity_threshold` | number | No | Minimum similarity score 0-1 (default: 0.3). Higher values return more relevant results. |
-| `project` | string | No | Project namespace to filter by (e.g.) |
+| `project` | string | No | Project namespace to filter by (e.g., 'my-project') |
========================================================================
# search
@@ -1877,11 +2094,11 @@ Initialize session, detect agent, load institutional context (last session, rece
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
-| `agent_identity` | `"CLI"` \| `"DAC"` \| `"CODA-1"` \| `"Brain_Local"` \| `"Brain_Cloud"` | No | Override agent identity (auto-detects if not provided) |
-| `linear_issue` | string | No | Current Linear issue identifier (e.g., OD-XXX) |
+| `agent_identity` | `"cli"` \| `"desktop"` \| `"autonomous"` \| `"local"` \| `"cloud"` | No | Override agent identity (auto-detects if not provided) |
+| `linear_issue` | string | No | Current Linear issue identifier (e.g., PROJ-123) |
| `issue_title` | string | No | Issue title for scar context |
| `issue_description` | string | No | Issue description for scar context |
| `issue_labels` | string[] | No | Issue labels for scar context |
| `project` | string | No | Project namespace (e.g., 'my-project'). Scopes sessions and searches. |
-| `force` | boolean | No | Force create new session even if one already exists (OD-558) |
+| `force` | boolean | No | Force create new session even if one already exists |
diff --git a/apps/docs/public/llms.txt b/apps/docs/public/llms.txt
index 59163a8..f6ece5e 100644
--- a/apps/docs/public/llms.txt
+++ b/apps/docs/public/llms.txt
@@ -8,7 +8,7 @@ Add GitMem to any MCP-compatible AI agent:
```bash
# Recommended: interactive setup (Claude Code, Cursor, etc.)
-cd your-project && npx gitmem init
+cd your-project && npx gitmem-mcp init
# Or add the MCP server directly
claude mcp add gitmem -- npx -y gitmem-mcp
@@ -56,7 +56,7 @@ Full docs: https://gitmem.ai/llms-full.txt
- [Your First Session](https://gitmem.ai/docs/getting-started/first-session): Walk through a complete GitMem session from start to close.
- [Free vs Pro](https://gitmem.ai/docs/getting-started/free-vs-pro): What you get today with the free tier and what's coming with Pro.
- [Overview](https://gitmem.ai/docs/getting-started): Get GitMem running in under 5 minutes.
-- [Installation](https://gitmem.ai/docs/getting-started/installation): Install GitMem via npx, global install, or MCP configuration.
+- [Installation](https://gitmem.ai/docs/getting-started/installation): Install GitMem via the init wizard, manual MCP config, or global install.
- [Closing Ceremony](https://gitmem.ai/docs/guides/closing-ceremony): How to properly close a GitMem session with reflection and persistence.
- [Error Handling](https://gitmem.ai/docs/guides/error-handling): Understanding and debugging GitMem errors.
- [Multi-Agent Workflows](https://gitmem.ai/docs/guides/multi-agent): Inject institutional memory into sub-agents and capture their findings.
@@ -73,7 +73,7 @@ Full docs: https://gitmem.ai/llms-full.txt
- [create_learning](https://gitmem.ai/docs/tools/create-learning): Create scar, win, or pattern entry in institutional memory.
- [create_thread](https://gitmem.ai/docs/tools/create-thread): Create an open thread to track unresolved work across sessions.
- [dismiss_suggestion](https://gitmem.ai/docs/tools/dismiss-suggestion): Dismiss a suggested thread.
-- [get_transcript](https://gitmem.ai/docs/tools/get-transcript): Retrieve a session transcript from storage (OD-467).
+- [get_transcript](https://gitmem.ai/docs/tools/get-transcript): Retrieve a session transcript from storage.
- [graph_traverse](https://gitmem.ai/docs/tools/graph-traverse): Traverse the knowledge graph over institutional memory triples.
- [health](https://gitmem.ai/docs/tools/health): Show write health for the current session.
- [gitmem-help](https://gitmem.ai/docs/tools/help): gitmem-help - Show available commands with ASCII art header.
@@ -86,7 +86,7 @@ Full docs: https://gitmem.ai/llms-full.txt
- [record_scar_usage_batch](https://gitmem.ai/docs/tools/record-scar-usage-batch): Track multiple scar applications in a single batch operation (reduces session close latency).
- [record_scar_usage](https://gitmem.ai/docs/tools/record-scar-usage): Track scar application for effectiveness measurement.
- [resolve_thread](https://gitmem.ai/docs/tools/resolve-thread): Mark an open thread as resolved.
-- [save_transcript](https://gitmem.ai/docs/tools/save-transcript): Save full session transcript to storage for training data and post-mortems (OD-467).
+- [save_transcript](https://gitmem.ai/docs/tools/save-transcript): Save full session transcript to storage for training data and post-mortems.
- [search_transcripts](https://gitmem.ai/docs/tools/search-transcripts): Semantic search over session transcript chunks.
- [search](https://gitmem.ai/docs/tools/search): Search institutional memory by query.
- [session_close](https://gitmem.ai/docs/tools/session-close): Persist session with compliance validation.