Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add OutlineDriven/odin-claude-plugin --skill repo-health-triagegit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/repo-health-triage)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/repo-health-triage"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/repo-health-triage/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/repo-health-triage"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/repo-health-triage.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00031 | $0.01117 |
| Opus 5 | $0.00015 | $0.00558 |
| Sonnet 5 | $0.00006 | $0.00223 |
| Haiku 4.5 | $0.00003 | $0.00112 |
Grade A, and why
repo-health-triage scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repo health triage
Contract
| Field | Bound contract |
|---|---|
| Trigger | A scheduled or explicitly requested repository-health pass spanning CI, pull requests, issues, commits, discussions, and durable run state. |
| Authority | Reversible local: writes only the bounded report file, one append-only run-log entry, and at most one isolated-fix proposal; rollback is deleting those files. No remote mutation. No source, label, merge, or close mutation without explicit approval. |
| Side effect | Local-write to a bounded High/Watch/Noise report plus one append-only run-log entry and at most one isolated-fix proposal; no unapproved source, label, merge, or close mutation. |
| Done | Every inspected signal lands in High, Watch, or Noise with an evidence line; the report is persisted or returned and any score stays informational, never a reason to act. |
Refusals
- Unapproved source, label, merge, or close mutation: rejected. No mutation without explicit approval.
- Numeric score as a reason to act: rejected. Any score is informational only and must not be cited as a reason to act.
- Scope widening into signal classes not requested or into mutation territory: rejected. Stop immediately and report the boundary violation.
Inputs
- Repository path or remote URL (required).
- Optional: scope filter limiting inspection to specific signal classes (CI, PRs, issues, commits, discussions, run state). If omitted, all signal classes are inspected.
- Optional: prior run-log path for append. If omitted, a new run-log is created.
Procedure
- Resolve the repository target and confirm read access. If access fails, stop and report the failure class. Done when: read access is confirmed or the failure class is reported.
- Enumerate open pull requests. For each, record: merge conflicts (High), CI red or no CI run (High), changes-requested review state (High), idle >14 days with no activity (Watch), and all others (Noise). Each classification carries an evidence line with the PR identifier and the specific signal. Done when: every open PR is classified with an evidence line.
- Enumerate open issues. For each, record: unanswered >7 days (Watch), linked CI failure (High), and all others (Noise). Each classification carries an evidence line. Done when: every open issue is classified with an evidence line.
- Enumerate recent commits on the default branch. For each, record: CI status red (High), CI status missing (Watch), and CI status green (Noise). Each classification carries an evidence line. Done when: every recent commit is classified with an evidence line.
- Enumerate discussions or forum threads if the repository platform supports them. For each, record: unanswered >7 days (Watch) and all others (Noise). Each classification carries an evidence line. Done when: every discussion is classified with an evidence line.
- Inspect durable run state (workflow runs, scheduled job status). For each, record: failed run (High), stale run with no recent execution (Watch), and healthy run (Noise). Each classification carries an evidence line. Done when: every run-state item is classified with an evidence line.
- Compile the bounded report: group all signals by classification (High, Watch, Noise). Each entry contains the signal source, identifier, evidence line, and classification. No signal appears in more than one bucket. Done when: the report is compiled with every signal in exactly one bucket.
- Append one entry to the run-log: timestamp, repository target, signal counts per classification, and report file path. Done when: the run-log entry is appended.
- If any High signal admits an isolated fix (typo in workflow file, missing CI config, stale label), propose exactly one fix with the target file, the proposed change, and the rationale. Do not apply the fix. If no isolated fix is available, skip this step. Done when: one fix proposal is made or the step is skipped.
- Return or persist the report. Any numeric score derived from the classification counts is informational only and must not be cited as a reason to act. Done when: the report is returned or persisted.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 52 lines · 31 tokens per session scan A 88c3532d6e46
repo-health-triage is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (36 stars, last pushed 3d ago), licensed Apache-2.0. It adds 31 tokens to every session and 1,117 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…