메뉴
Use when building or refreshing the Cartographer per-repo atlas, a small codebase map of layers, naming conventions, error style, and extension points. Triggers on "/survey", "survey repo", "refresh atlas", "rebuild codebase map".
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
Use when designing a goal-oriented agent loop or reviewing one for runaway risk - build gate, machine-decidable goal, anti-Goodhart boundaries, five failure modes. Triggers on "design a loop", "review this loop", "will this loop run away".
Use when running OpenAI Codex via the codex CLI (copilot CLI fallback) for code analysis, refactoring, or editing. Triggers on "run codex", "codex analyze", "ask codex", "copilot with codex".
Use when running Google Gemini via the native Gemini CLI for code analysis, refactoring, or editing. Triggers on "run gemini", "gemini analyze", "ask gemini".
Use when running a local Qwen model (via llama.cpp and the pi agent) for code analysis, refactoring, or editing - local inference, no API cost. Triggers on "run qwen", "qwen analyze", "ask qwen", "local model task".
Use when running Anthropic Claude Sonnet via the native claude CLI (headless) for code analysis, refactoring, or editing. Triggers on "run sonnet", "sonnet analyze", "ask sonnet".
Use when converting a plan, design doc, brainstorming output, or discovery doc into a structured PRD saved to the backlog. Triggers on "create PRD", "create PRD from plan", "make PRD", "convert to PRD", "save plan as PRD".
| name | survey |
| description | Use when building or refreshing the Cartographer per-repo atlas, a small codebase map of layers, naming conventions, error style, and extension points. Triggers on "/survey", "survey repo", "refresh atlas", "rebuild codebase map". |
Build or refresh the Cartographer atlas for the current repository — a proactive map of "where things live" that downstream phases (Recon Brief, Architect, Conformance) consult before deciding where to make a change.
The atlas lives at ~/.claude/cartographer/projects/<hash>/:
atlas.json — machine-readable summary (layers, naming, error style,
dependency edges, forbidden imports, staleness config).atlas.md — human-readable 2-5KB summary.staleness.flag — empty marker; present means the atlas has drifted.atlas.json exists and staleness.flag is absent
(prints a skip reason). Runs the survey otherwise (atlas missing or stale).--if-missing — Survey only when atlas.json is absent. No-op when an atlas
already exists, even if staleness.flag is present (a stale atlas is left
for the bare invocation or --refresh to rebuild). Used by /catchup, which
only needs an atlas to exist.--refresh — Force a full re-survey, bypassing the staleness check. Clears
staleness.flag on success. Preserves any [manual] override block in
atlas.json byte-for-byte; rewrites every other field.Delegate to the survey script, forwarding any flag verbatim:
python3 "${CLAUDE_SKILL_DIR}/scripts/run.py"
python3 "${CLAUDE_SKILL_DIR}/scripts/run.py" --if-missing
python3 "${CLAUDE_SKILL_DIR}/scripts/run.py" --refresh
The script resolves the project hash via _lib_cartographer, walks the repo
tree, extracts symbols (tree-sitter, with a regex fallback marked degraded),
classifies layers, naming, and error style, then writes atlas.json and
atlas.md atomically.
Report the status line the script emits — whether it surveyed, refreshed, or skipped, plus the atlas location and size. If the script reports a degraded run (tree-sitter unavailable) or a truncated atlas, surface that to the user.
~/.claude/cartographer/projects/<hash>/.[manual] block in atlas.json lets the
user override it and survives re-survey.truncated: true.head_sha and staleness checks skip.