Borrowing it
Nothing to install: this file belongs to xsovad06/sova. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/xsovad06/sova/main/.claude/commands/health-audit.mdgit clone --depth 1 https://github.com/xsovad06/sovaWrote 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/commands/xsovad06/sova/health-audit)<a href="https://agentmods.dev/commands/xsovad06/sova/health-audit"><img src="https://agentmods.dev/badge/commands/xsovad06/sova/health-audit/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/commands/xsovad06/sova/health-audit"><img src="https://agentmods.dev/badge/commands/xsovad06/sova/health-audit.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 | $0.00024 | $0.11558 |
| Opus 5.5 | $0.00010 | $0.04623 |
| Sonnet 5.5 | $0.00005 | $0.02312 |
| Haiku 4.5 | $0.00002 | $0.01156 |
Grade A, and why
health-audit scanned grade A with 1 finding 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.
Reads agent configuration directorieslowAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
find . -name "*.py" -not -path "./.venv/*" -not -path "*/__pycache__/*" -not -path "*/migrations/*" -not -path "./.claude/*" | xargs wc -l | tail -1 Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 949 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deep Technical Health Audit
Focus area: $ARGUMENTS
Role
You are the original architect and sole developer of this application, a senior software engineer with 20 years of experience who wrote every line from scratch. You know every design decision, every shortcut, every deferred TODO. You have been asked to produce an honest, thorough technical health report as if preparing the project for: (a) onboarding a new senior engineer, (b) a due-diligence review, and (c) your own prioritized improvement backlog.
Be brutally honest. Flag what is genuinely good (so it is preserved), and be specific about what is weak (so it can be fixed). Avoid generic advice. Every finding must reference concrete code, files, or patterns in THIS codebase.
Procedure
Step 1: Staleness check
Report path confidentiality: before anything else, and regardless of
whether docs/HEALTH-AUDIT.md exists yet, check whether writing to it is
safe. A "withhold" or "file neutrally" disclosure choice (Phase C)
deliberately keeps the full exploit path in this file's body: publishing it
through a commit instead of an issue defeats the point of withholding it,
and the gate this check feeds (Phase C's unsafe-path block) has nothing to
consult without it, which is exactly the portable, first-run case this check
exists for. This repo gitignores the file, but the distributable copy of
this command runs in arbitrary projects where docs/ is normally tracked.
The check below works whether or not the file exists yet (an absent path is
simply untracked and not ignored, unless docs/ itself is gitignored), so it
does not depend on the "does the file already exist" check further below.
if git ls-files --error-unmatch docs/HEALTH-AUDIT.md >/dev/null 2>&1; then
echo "tracked"
elif git check-ignore -q docs/HEALTH-AUDIT.md; then
echo "ignored"
elif git rev-parse --is-inside-work-tree >/dev/null 2>&1; then
echo "untracked-and-not-ignored"
else
echo "not-a-git-repo"
fi
ignored: safe, no action needed.untracked-and-not-ignored: this is the default state of a fresh project (no report file yet, and nothing ignoringdocs/), so do not treat it as a hard stop here: it would stop the audit before Step 5's expensive fan-out and Step 6's claims ledger have even run, regardless of whether any finding this run turns out to be security-sensitive. Warn the user now that any future withheld or neutrally-filed exploit detail would land in a file git can commit, and suggest addingdocs/HEALTH-AUDIT.mdto.gitignorebefore continuing. Record the status as unsafe for Phase C, which enforces the actual block, and only for a group whose disclosure choice turns out to need it.tracked: same deferred treatment as above, record unsafe, except the eventual fix is not optional: if a group's choice needs confidentiality, the file must also be removed from version control, not just gitignored going forward (its history would still carry the committed content).not-a-git-repo(not inside a git working tree): state plainly thatdocs/HEALTH-AUDIT.mdhas no git protection in this project at all (there is no.gitignoreto edit). Record unsafe; Phase C's block, if it fires, must tell the user to pick a different, actually-protected path for the report rather than prescribing a.gitignoreedit.
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 Changed cecc263f6a2a
- 13d ago Changed · +497 lines 82f98dd391a8
- 16d ago Changed a0495aa56425
- 1mo ago Changed · +16 lines 1fc9e2bacdcc
- 1mo ago Changed · -19 lines 441725ce781a
- 2mo ago First seen · 455 lines · 24 tokens per session scan A fa050a898365
health-audit is a command published in the GitHub repository xsovad06/sova (2 stars, last pushed yesterday), licensed Apache-2.0. It adds 24 tokens to every session and 11,558 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
greptile-review
Execute the full Greptile review cycle on a PR, looping until score meets threshold or circuit breaker triggers.
review-branch
Review the current branch's diff against base by dispatching atomic-reviewer. No orchestration loop, no spec required — pre-flight before /commit pr or /commit merge.
advanced-code-review-context
Advanced Code Review Phase 2: Context Analysis - load previous reviews, PR history, declined items.
code-review-give
Give mode for code-review: Review someone else's code with multi-pass analysis and structured recommendations.
mach12:issue-review
Read a GitHub issue and all comments, review the implementation plan, and present findings.
review-design-verify
Phases 4-5 of reviewing-design-docs: Interface Verification + Implementation Simulation.