메뉴
This skill should be used when the user asks to "apply insights", "insights to rules", "update rules from insights", "apply suggestions", or mentions updating CLAUDE.md from insights data. Extracts suggestions from the /insights report and applies them to the global CLAUDE.md via the sync pipeline.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | insights-apply |
| description | This skill should be used when the user asks to "apply insights", "insights to rules", "update rules from insights", "apply suggestions", or mentions updating CLAUDE.md from insights data. Extracts suggestions from the /insights report and applies them to the global CLAUDE.md via the sync pipeline. |
Extract CLAUDE.md improvement suggestions from the /insights report and apply them through the sync pipeline. Targets the global CLAUDE.md only (insights are cross-project).
/insights and seeing claude_md_additions suggestions~/.claude/usage-data/~/.claude/usage-data/report.html or facets data)~/.claude/CLAUDE.mdExecute these steps in order.
Locate and parse the insights data:
# Check for insights data
if [ -f ~/.claude/usage-data/report.html ]; then
echo "Found insights report"
elif [ -d ~/.claude/usage-data/facets/ ]; then
echo "Found facets data"
else
echo "No insights data found. Run /insights first."
exit 1
fi
Extract the claude_md_additions array from the report. These are structured suggestions with:
title: Section heading for CLAUDE.mdcontent: The actual rules/guidance to addsource: Which sessions/patterns generated this suggestionpriority: How impactful the suggestion is (high/medium/low)Show each suggestion with context for user review:
┌───────────────────────────────────────────────────────────────┐
│ INSIGHTS SUGGESTIONS (N found) │
├───────────────────────────────────────────────────────────────┤
│ │
│ Suggestion 1/N: <title> │
│ Source: <N sessions with this pattern> │
│ Priority: <high|medium|low> │
│ │
│ Proposed CLAUDE.md addition: │
│ ┌─────────────────────────────────────────────────────────┐ │
│ │ ## <Title> │ │
│ │ - <rule 1> │ │
│ │ - <rule 2> │ │
│ │ - <rule 3> │ │
│ └─────────────────────────────────────────────────────────┘ │
│ │
└───────────────────────────────────────────────────────────────┘
For each suggestion, ask the user:
{
"questions": [{
"question": "Apply this suggestion to global CLAUDE.md?",
"header": "Apply",
"multiSelect": false,
"options": [
{
"label": "Apply (Recommended)",
"description": "Add this section to ~/.claude/CLAUDE.md"
},
{
"label": "Skip",
"description": "Don't add this suggestion"
},
{
"label": "Edit first",
"description": "Let me modify the content before applying"
}
]
}]
}
For each approved suggestion, use the CLAUDE.md sync pipeline:
# Apply using the sync pipeline
python3 utils/claude_md_sync.py ~/.claude/CLAUDE.md --add-section "<title>" "<content>"
If sync pipeline is not available, fall back to direct append:
# Fallback: Direct append to CLAUDE.md
echo "" >> ~/.claude/CLAUDE.md
echo "## <Title>" >> ~/.claude/CLAUDE.md
echo "" >> ~/.claude/CLAUDE.md
echo "<content>" >> ~/.claude/CLAUDE.md
After applying all approved suggestions, verify the CLAUDE.md stays within budget:
# Run budget enforcer
python3 utils/claude_md_optimizer.py ~/.claude/CLAUDE.md --check
# If over budget, the optimizer will:
# 1. Report which sections exceed limits
# 2. Suggest P2 content to move to detail files
# 3. Offer to auto-optimize
If the CLAUDE.md exceeds the line budget after additions:
┌───────────────────────────────────────────────────────────────┐
│ ⚠️ BUDGET WARNING │
├───────────────────────────────────────────────────────────────┤
│ CLAUDE.md is now N lines (budget: 200 lines) │
│ Over budget by M lines │
│ │
│ Options: │
│ 1. Run optimizer to trim P2 content │
│ 2. Keep as-is (will be truncated at 200 lines) │
│ 3. Manually edit to reduce size │
└───────────────────────────────────────────────────────────────┘
Summarize what was applied:
┌───────────────────────────────────────────────────────────────┐
│ INSIGHTS APPLY REPORT │
├───────────────────────────────────────────────────────────────┤
│ Suggestions found: N │
│ Applied: M │
│ Skipped: P │
│ Edited: Q │
│ │
│ CLAUDE.md status: Within budget / Over budget │
│ File: ~/.claude/CLAUDE.md │
│ │
│ Applied sections: │
│ ✅ Branch Guard & Git Hooks │
│ ✅ Worktree Best Practices │
│ ⏭️ Release Pipeline (skipped) │
│ ✅ Testing Conventions (edited) │
└───────────────────────────────────────────────────────────────┘
Use craft box-drawing format throughout. Show step progress:
[1/5] Parse insights ........... DONE (N suggestions found)
[2/5] Present suggestions ...... DONE (user reviewed N)
[3/5] Apply approved ........... DONE (M applied)
[4/5] Budget check ............. DONE (within budget)
[5/5] Report ................... SHOWN
| Error | Recovery |
|---|---|
| No insights data | Suggest running /insights first |
| CLAUDE.md not found | Suggest the docs/claude-md skill (init) |
| Sync pipeline fails | Fall back to direct append |
| Over budget | Offer optimizer or manual edit |
| Parse error | Show raw data for manual review |
Global CLAUDE.md only. Insights are cross-project patterns, so they belong in ~/.claude/CLAUDE.md, not project-specific files.
/insights — Generate the insights reportutils/claude_md_sync.py — 4-phase sync pipelineutils/claude_md_optimizer.py — Budget enforcementThis skill should be used when the user asks to "pre-flight check", "validate before commit", "validate before PR", "validate before deploy", "is this ready to ship", "check readiness", "run pre-flight", "validate project state", or wants context-aware validation across commit/PR/release/deploy boundaries. Orchestrates universal validators (lint, tests, types, version sync, stale refs, docs, hook conflicts) and supports generating new custom validators.
This skill should be used when a command's --refine flag is set, or the user asks to "refine my prompt", "optimize this prompt", "make this request sharper" — rewrites a vague natural-language request into a specific, well-structured prompt using project context, shows before/after, and confirms before the caller proceeds. Replaces the deprecated /craft:refine command.
This skill should be used when the user asks to "release", "create a release", "ship it", "publish a new version", "bump version and release", "cut a release", or mentions releasing to GitHub/npm/PyPI. Orchestrates the full release pipeline from pre-flight checks through GitHub release creation.
ADHD-friendly session and task workflow for craft users — captures context at session end, restores it at session start, picks the next task, breaks through blockers, holds focus on one task, reviews a craft spec, or refines a vague craft prompt. Use when wrapping up or resuming a coding session, when deciding what to work on next (also "what's next", "pick a task for me", "give me something to do"), when stuck or in decision paralysis, when focusing on a single craft task, or when reviewing/refining a craft spec or prompt. The full intent-to-operation map is in the "When to Use" table below.
This skill should be used when the user asks to "init a repo", "set up git", "manage branches", "switch branch", "create a worktree", "list worktrees", "clean merged branches", "sync with remote", "git activity", "git recap", "protect main", "apply branch protection", "bypass protection", "undo a commit", "fix a git mistake", or mentions git lifecycle, worktrees, branch protection, GitHub-side baseline protection, safety rails, or learning git workflows. Covers craft's full git lifecycle — repo init, branch and worktree management, remote sync, status/recap, local and GitHub-side branch protection, and the learning/safety reference docs.
This skill should be used when the user asks "what's my friction", "session insights", "friction report", "what patterns am I hitting", "analyze my sessions", "generate insights report", "aggregate facets", or "show CLAUDE.md suggestions". Aggregates session facet history into a friction/goals report. Formerly bundled with the brainstorm skill (ideation); split out because the two operations share no input or output. Directory name kept as brainstorm-insights for path stability — tests and references/insights.md reference it.