mine-history
Extract and synthesize learnings from past Claude Code session transcripts into project documentation. Use when asked to "mine history", "extract learnings from past sessions", "what patterns keep recurring", or "improve docs from history".
What this skill does
# Mine Session History
Extract actionable learnings from past Claude Code conversations and synthesize them into project documentation. Turns the goldmine of corrections, mandates, and discoveries across all sessions into persistent knowledge.
$ARGUMENTS
## Phase 1: Extract Raw Corrections
Run the extraction script to scan all session transcripts for user corrections:
```bash
node scripts/extract-learnings.js \
--project=. \
--output=/tmp/learnings-raw.md \
--since-days=90 \
--max-pairs=500
```
Adjust `--since-days` and `--max-pairs` based on user arguments.
Read the output file to understand what was extracted.
## Phase 2: Audit Existing Documentation
Before proposing any changes, read ALL existing documentation to understand what's already captured — CLAUDE.md, docs/*.md, .claude/rules/. Build a mental map of what's already documented.
## Phase 3: Synthesize & Deduplicate
Work through the extracted corrections by topic:
| Check | Action |
|-------|--------|
| Already documented? | Skip — note as "confirmed" |
| Contradicts existing docs? | Flag for review |
| Genuinely new pattern? | Propose addition |
| Session-specific (not generalizable)? | Skip |
| Single occurrence (might be wrong)? | Skip unless obviously correct |
**Quality filter — only propose learnings that:**
- Appear multiple times across sessions (pattern, not one-off)
- OR are clearly universal (e.g., "never use DateTime.UtcNow")
- Would prevent a real mistake if documented
- Are concise enough for a one-liner or short paragraph
**Route by category:**
| Category | Target File |
|----------|------------|
| Universal rule | CLAUDE.md (if under 300 lines) |
| Domain-specific pattern | Appropriate docs/*.md file |
| Path-scoped convention | .claude/rules/{topic}.md |
| Tool/workflow preference | CLAUDE.md or docs/DEVELOPMENT_PRACTICES.md |
## Phase 4: Present Proposals
Present ALL proposed changes in a single summary before writing anything:
```markdown
## History Mining Results
### Statistics
- Sessions scanned: X
- Corrections found: Y
- Already documented: Z
- New learnings: N
### Proposed Updates
#### 1. docs/FRONTEND.md
- **Add to Reactivity section**: "Never unwrap .value in addToast() calls — refs stay reactive"
#### 2. Skipped (already documented)
- "Use npm run scripts" — in CLAUDE.md line 12
```
## Phase 5: Apply with Approval
Use AskUserQuestion to get approval before writing. Options: apply all, let me pick, save proposals to file, or skip.
After applying, show a summary table with file, change description, and line delta.
## Gotchas
- Never add session-specific context — "We fixed bug #42" is not a learning
- Prefer updating over adding — extend existing bullets, don't create new sections
- One-liners only for CLAUDE.md — it has a 300-line budget
- Route domain knowledge to docs/*.md — that's where detailed patterns belong
- Don't duplicate linter rules — if the toolchain enforces it, don't document it
- Multiple occurrences = high confidence — a pattern seen across 3+ sessions is almost certainly worth documenting
- Single occurrence = verify first — read the surrounding context before proposing
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.