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 davesheffer/hunch --skill healgit clone --depth 1 https://github.com/davesheffer/hunchWrote 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/davesheffer/hunch/heal)<a href="https://agentmods.dev/skills/davesheffer/hunch/heal"><img src="https://agentmods.dev/badge/skills/davesheffer/hunch/heal/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/davesheffer/hunch/heal"><img src="https://agentmods.dev/badge/skills/davesheffer/hunch/heal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00021 | $0.00228 |
| Opus 5.5 | $0.00008 | $0.00091 |
| Sonnet 5.5 | $0.00004 | $0.00046 |
| Haiku 4.5 | $0.00002 | $0.00023 |
Grade A, and why
heal 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 1mo 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.
What it actually says
Reconcile decision-grounding drift for $ARGUMENTS (or the whole repo).
- Run
hunch drift(orhunch heal) to list doc≠graph anchor-stale sections — a file still anchored to a superseded decision while a current one exists. Only explicit topic anchors fire; never a semantic guess. - For each, assume the DOC is stale first (Heal A). Propose an edit bringing the file to the CURRENT decision; show it as a diff and wait for my confirm. Never rewrite prose silently.
- Only if I explicitly say "the DECISION is stale, not the doc" (Heal B): run /hunch:capture to record a superseding decision, then return to step 2 — the prose re-derives from the new decision as a separate confirm.
- Report: healed (Heal A), superseded (Heal B), skipped. Never touch the graph except via an explicit Heal B capture.
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.
- 1mo ago First seen · 10 lines · 21 tokens per session scan A e6de4bea487e
heal is a skill published in the GitHub repository davesheffer/hunch (11 stars, last pushed 3d ago), licensed Apache-2.0. It adds 21 tokens to every session and 228 once invoked, about $0.0001 per session on Opus 5.5. 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 skills, from other repositories
code-review
Orchestrate an adaptive multi-agent review of a pull request or local change that builds a structural and semantic brief, assigns risk-based reviewer lanes at runtime, challenges and deduplicates their findings, verifies each one, and renders an evidence-backed review. Use when a pull request or local change needs…
tech-debt-audit
Audit a whole repository or subsystem for technical debt with at most four non-overlapping read-only subagents, lead verification, and interest-over-principal ranking, and produce an evidence-backed debt register. Use when the user wants to find, size, or prioritize existing debt across a codebase; not for reviewing…
changes-report
Report all repository changes that entered a target branch during one required date or period, grouped by product change and written as plain English for a broad audience. Use for daily, weekly, date-range, or relative-period change summaries rather than explaining one known diff.
jira-issue-create
Create one or more Jira issues through acli from user-approved drafts using heading-based Context, Acceptance criteria, and Engineering notes sections. Use only when explicitly invoked to create Jira issues; use jira-issue-refine when requirements still need refinement.
coderabbit-triage
Triage unresolved CodeRabbit feedback on the current GitHub pull request. Use for read-only disposition and, only when separately authorized, scoped remediation and remote thread closure.
claude-review
Run an independent Claude Opus review of the complete current change against the repository's default branch, then verify its findings locally. Use after focused checks pass when you want a second model to review the change before further edits or delivery.