← ClaudeAtlas

index-reposlisted

Use this skill when expanding a small set of candidate skills into the full breadth of what their host repos offer. Enumerates the full content of one or more GitHub repos that host Claude Code skills, agents, and plugins. Thin wrapper around `scripts/scan_repo.mjs` (deterministic Node scanner — no LLM required). Returns structured markdown indices per repo.
ievo-ai/skills · ★ 0 · AI & Automation · score 72
Install: claude install-skill ievo-ai/skills
# Index Repos Enumerate GitHub repo content. **v0.4.0 architectural change:** delegates all heavy lifting to `scripts/scan_repo.mjs` — a pure-Node deterministic scanner that the `ievo-ai/community-index` GitHub Action ALSO uses. Single source of truth for the indexing algorithm. (v0.3.x used Python; migrated to Node so no extra runtime install needed — Node ships with Claude Code.) ## Input Either: - A list of `<owner>/<repo>` strings - A single repo identifier Caller provides via prompt. No interactive input. ## What scan_repo.mjs does (executed via Bash) For each repo: 1. Shallow clone into `~/.ievo/checkouts/<owner>-<repo>-<hash>/` (a content hash suffix keeps two different repos from colliding on the same cache directory; refreshes if cached + TTL expired, verifying the cached checkout's git remote still matches before reusing it) 2. Detect layout (marketplace / flat-skills / flat-agents / single-plugin / other) 3. Enumerate plugins / agents / skills / commands / hooks / MCP 4. Parse YAML frontmatter for each item 5. Compute risk signals (hooks events, allowed-tools, license, etc.) 6. Emit structural facts only — NO risk_tier (removed in v0.5.2). Risk verdicts come from `security-auditor` antivirus deep scan per selected item before install. 7. Write `<owner>-<repo>-<hash>.md` (same hash-suffixed naming as the checkout dir above, so a colliding slug can't overwrite another repo's index) matching `ievo-ai/community-index/indices/_TEMPLATE.md` format 8. Write `<owner>