一键在 Manus 中运行任何 Skill

开始使用

repo-map

星标2

分支0

更新时间2026年5月17日 11:29

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

agentculture

agentculture/code-lens-cli

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

repo-map

Three verbs, scaling from cheap (one repo, mechanical facts) to opt-in expensive (entire connected component with deep mind-model materials):

Mode	Invocation
One repo, mechanical facts	`scripts/profile.sh <path>`
One repo, mind-model materials	`scripts/profile.sh <path> --depth deep`
Walk N hops from a repo	`scripts/connections.sh <path> --depth N`
Whole workspace at once	`scripts/graph.sh [<root>...]`

When to use

When the user asks about a repo or a set of related repos: profile, deps, who-cites-whom, who-vendors-whom. The scripts collect the deterministic facts so you (the agent) can spend your tokens on synthesis, not grep.

Token math: one profile.sh call replaces ~6–10 separate Read calls (README.md, pyproject.toml, CHANGELOG.md, CLAUDE.md, docs/skill-sources.md, CITATION.md, culture.yaml, plus directory listings for the package tree and vendored skills). If your task needs ≥2 of those, call profile.sh once instead of reading them piecemeal.

When NOT to use

You already know the exact file you need to read (e.g. one specific function's source) — Read it directly.
You need to modify the repo — repo-map is read-only.
The repo isn't a Python project (no pyproject.toml) — profile degrades but the mechanical-facts payload shrinks to mostly empty.

Compose recipe

For the question "explain how these repos fit together":

scripts/graph.sh [<root>] once — gets every repo + every edge.
If a particular repo needs depth, scripts/profile.sh <path> --depth deep.
Synthesize the narrative from those outputs. Do not re-grep what profile already reports.

For the question "tell me about this repo and what it connects to":

scripts/profile.sh <path> (or --depth deep for narrative-rich).
scripts/connections.sh <path> --depth 1 --profile for immediate neighbors; raise --depth to widen.

Output

Markdown by default — read it directly. Errors include reason: and hint: lines. During walks, per-node errors are inlined and the walk continues (pass --strict to flip).

Configuration

Optional .claude/skills/repo-map/config.json for per-workspace defaults:

{
  "roots": ["$HOME/git"],
  "additional_markers": ["culture.yaml"],
  "skip_dirs": [".git", ".venv", "node_modules", "__pycache__"],
  "default_connections_depth": 1
}

Flags always override config.

Engine

The actual logic lives in code_lens/repo/ and is invoked via uv run python -m code_lens.repo <verb>. The shell scripts are one-line wrappers; the agent-facing contract is the verbs and their flags, not the wrappers.

Interpreter note: the scripts use uv run --directory <project-root> so they work regardless of the caller's working directory. uv resolves the correct virtualenv and lock file from the project root automatically.

同仓库更多 Skills

同仓库

recall

agentculture/code-lens-cli

Search the shared eidetic memory store and get back ranked, provenanced records. Drives `eidetic recall` with four search modes — exact (verbatim substring), approximate (vector/semantic), keyword (BM25 lexical), and hybrid (a weighted blend of vector+keyword, the default) — each hit carrying its text, full metadata, a relevance `score`, and a freshness `signal`. Recall passively reinforces matched records (bumps last_recall + recall_count). Shadowed and archived records are excluded by default; use --include-shadowed / --include-archived to retrieve them. The store lives at ~/.eidetic/memory (a home-dir path outside any git worktree); the wrapper defaults queries to this agent's PERSONAL, PRIVATE scope (`--scope code-lens-cli --visibility private`, suffix read from culture.yaml) — matching where /remember writes — so a no-flag recall returns this agent's own private records plus the shared public pool, and Claude and the colleague backend recall each other's memories because both resolve the same suffix via

2026-06-232

remember

agentculture/code-lens-cli

Ingest records into the shared eidetic memory store so they can be recalled later. Drives `eidetic remember`: accepts one record as a JSON object, or a batch as NDJSON on stdin for bulk ingest. Upsert is idempotent by id (and dedups by content hash) — re-remembering updates in place, never duplicates. Stamps a `created` date on every record at ingest time. Accepts `supersedes` (id of the record this one replaces, for within-scope shadowing via `sweep`) and `links` (list of related-memory ids). The store lives at ~/.eidetic/memory (a home-dir path outside any git worktree), and the wrapper defaults records to this agent's PERSONAL, PRIVATE scope (`--scope code-lens-cli --visibility private`, suffix read from culture.yaml) so they don't leak to a default/other-scope recall — Claude and the colleague backend still share them because both resolve the same suffix via this skill. Pass `--visibility public` to contribute to the shared public pool instead. Use when the user says "remember this", "store this", "save t

2026-06-232

cicd

agentculture/code-lens-cli

Steward's CI/CD lane, layered on `agex pr`. Delegates lint / open / read / reply / delta to agex; adds two steward extensions — `status` (SonarCloud quality gate + hotspots + unresolved-thread tally) and `await` (read --wait + status with non-zero exit on Sonar ERROR or unresolved threads). Use when: creating PRs in steward, handling review feedback, polling CI status, or the user says "create PR", "review comments", "address feedback", "resolve threads". Renamed from `pr-review` in steward 0.7.0; rebased on agex in 0.12.0.

2026-05-172

code-lookup

agentculture/code-lens-cli

Pick the verb by the question you're asking, not the tool you'd otherwise reach for. Three questions, three one-call answers: **"What changed in the last N commits — which functions or classes did each commit add, remove, or modify?"** → use `recent`, **not** `git log` + `git show` + manual diff reading. `scripts/recent.sh [<path>] [-n N]` returns `{"commits": [{"sha","date","subject","changes":[{"file","added","removed","modified"}, …]}, …]}` — the symbol diff is computed via AST, so you skip the manual "what functions changed" step. The `modified` heuristic compares function line-spans; pure line shifts can cause false positives (a deferred improvement). **"Where is `<pattern>` referenced — and what function or class owns each match?"** → use `grep`, **not** `rg`/`grep` + Read each file to find the enclosing scope. `scripts/grep.sh <pattern> [<path>]` returns `{"pattern","matches":[{"file","line","scope","text"}, …]}` — `scope` is the enclosing Python function or class, resolved via AST. Requires `ripgrep`

2026-05-172

communicate

agentculture/code-lens-cli

Cross-repo + mesh communication from steward: file tracked GitHub issues on sibling repos, fetch issues from sibling repos to inline current state into briefs, and send live messages to Culture mesh channels. Use when the next step lives outside steward (a brief for a sibling-repo agent, a status ping for a Culture channel, or pulling an issue body + comments into context). Issue posts auto-sign with `- steward (Claude)`; mesh messages are unsigned (the IRC nick is the speaker). Not for in-steward issues — use `gh issue create` or the `cicd` skill for those. Renamed from `coordinate` in steward 0.8.0; absorbed `gh-issues` in 0.9.1. Issue I/O is backed by `agtag` (>=0.1) starting in 0.11.0.

2026-05-172

run-tests

agentculture/code-lens-cli

Run pytest with parallel execution and coverage. Use when running tests, verifying changes, or the user says "run tests", "test", or "pytest".

2026-05-172

name

repo-map

description

repo-map

Three verbs, scaling from cheap (one repo, mechanical facts) to opt-in expensive (entire connected component with deep mind-model materials):

Mode	Invocation
One repo, mechanical facts	`scripts/profile.sh <path>`
One repo, mind-model materials	`scripts/profile.sh <path> --depth deep`
Walk N hops from a repo	`scripts/connections.sh <path> --depth N`
Whole workspace at once	`scripts/graph.sh [<root>...]`

When to use

When NOT to use

You already know the exact file you need to read (e.g. one specific function's source) — Read it directly.
You need to modify the repo — repo-map is read-only.
The repo isn't a Python project (no pyproject.toml) — profile degrades but the mechanical-facts payload shrinks to mostly empty.

Compose recipe

For the question "explain how these repos fit together":

scripts/graph.sh [<root>] once — gets every repo + every edge.
If a particular repo needs depth, scripts/profile.sh <path> --depth deep.
Synthesize the narrative from those outputs. Do not re-grep what profile already reports.

For the question "tell me about this repo and what it connects to":

scripts/profile.sh <path> (or --depth deep for narrative-rich).
scripts/connections.sh <path> --depth 1 --profile for immediate neighbors; raise --depth to widen.

Output

Markdown by default — read it directly. Errors include reason: and hint: lines. During walks, per-node errors are inlined and the walk continues (pass --strict to flip).

Configuration

Optional .claude/skills/repo-map/config.json for per-workspace defaults:

{
  "roots": ["$HOME/git"],
  "additional_markers": ["culture.yaml"],
  "skip_dirs": [".git", ".venv", "node_modules", "__pycache__"],
  "default_connections_depth": 1
}

Flags always override config.

Engine

Interpreter note: the scripts use uv run --directory <project-root> so they work regardless of the caller's working directory. uv resolves the correct virtualenv and lock file from the project root automatically.