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 agentmods add commands/sectiontn/stale-docs/stale-docsgit clone --depth 1 https://github.com/SectionTN/stale-docsWhat 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 | $0.00008 | $0.00342 |
| Opus 5 | $0.00004 | $0.00171 |
| Sonnet 5 | $0.00002 | $0.00068 |
| Haiku 4.5 | $0.00001 | $0.00034 |
Grade A, and why
stale-docs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run a full-repo staleness audit and report the results.
- Run the scanner in audit mode from the repository root:
node "${CLAUDE_PLUGIN_ROOT}/hooks/check-stale.js" --audit
-
The output is a ranked list capped at the 10 most confident findings, in the form
N. doc.md:line mentions \x` (fenced code block), from src/file.js`. Findings in fenced code blocks or referencing file paths rank above prose mentions. -
For each finding, read the doc location and the referenced source file, then judge whether the doc still matches the code (signature, parameters, flags, defaults, behavior). Judge against the source only. Never mark a claim accurate because it agrees with another doc or sounds right; the scanner reports references, not verdicts, and a reference can be perfectly accurate.
-
Present the results as a table with three columns:
file:line, what it references, and a verdict. Use stale when the doc contradicts current code (say what differs), accurate when you verified it against source, and unclear when it needs a human. -
If any findings are stale, offer to repair them with the stale-docs skill: patch wrong details, delete claims about code that no longer exists, rewrite a file from the source when most of it is stale.
If the scanner prints no stale doc references found or no doc files found, say so and stop.
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.
- yesterday First seen · 25 lines · 8 tokens per session scan A 7d3dddabe189
stale-docs is a command published in the GitHub repository SectionTN/stale-docs (3 stars, last pushed 1mo ago), licensed MIT. It adds 8 tokens to every session and 342 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
review-renovate
Review and merge renovate PRs with automerge configuration updates.
monitor-ci
Monitors pull request CI checks until they are resolved (pass or fail).
plan-regression-tests
Plan regression tests for existing code with it.skip statements.
ado-pull
Pull latest changes from Azure DevOps (like git pull). Supports increment, project, or full living docs sync.
github-push
Push local progress to GitHub Issues (like git push). Updates issue with task completion checklist.
save
Save this conversation as a new or existing reusable context.