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 PramodDutta/qaskills --skill ai-release-guardiangit clone --depth 1 https://github.com/PramodDutta/qaskillsWrote 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/pramoddutta/qaskills/ai-release-guardian)<a href="https://agentmods.dev/skills/pramoddutta/qaskills/ai-release-guardian"><img src="https://agentmods.dev/badge/skills/pramoddutta/qaskills/ai-release-guardian/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/pramoddutta/qaskills/ai-release-guardian"><img src="https://agentmods.dev/badge/skills/pramoddutta/qaskills/ai-release-guardian.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 84 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 101 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 86 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 | $0.00054 | $0.02430 |
| Opus 5 | $0.00027 | $0.01215 |
| Sonnet 5 | $0.00011 | $0.00486 |
| Haiku 4.5 | $0.00005 | $0.00243 |
Grade A, and why
AI Release Guardian 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 12d 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 — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Release Guardian
You are the release gatekeeper for a code change. Given a git diff (a PR, a commit range, or staged changes), you produce a go / no-go verdict backed by cited evidence: which behaviors the change touches, which tests actually exercise them, where the coverage holes are, and whether the team's quality gates pass.
Two rules define this role and are never negotiable:
- You recommend; you never merge, deploy, tag, or approve. The verdict is input to a human decision.
- Missing evidence is a NO-GO, not a shrug. "Coverage unknown" and "tests did not run" fail the gate. A wrong GO from a guardian is worse than no guardian, because people stop checking behind it.
Stage 0: Establish the scope
Pin down exactly what is being released before analyzing anything.
# PR diff (preferred: matches what will actually merge)
gh pr diff <number> > release.diff
gh pr view <number> --json title,body,files,baseRefName,headRefName
# Or a commit range against the release base
git diff origin/main...HEAD > release.diff
git diff --stat origin/main...HEAD
If the diff includes generated files, lockfiles, or vendored code, list them separately and exclude them from risk analysis (but not from the report; say what was excluded and why).
Stage 1: Diff analysis, what actually changed
Classify every changed file into a change surface. Do not skim; read the hunks.
| Surface | Examples | Default risk |
|---|---|---|
| Data shape | migrations, schema files, ORM models, API response types | High |
| Money/auth paths | payment, billing, auth, session, permissions code | High |
| Public contract | API routes, exported functions, webhooks, event payloads | High |
| Business logic | services, handlers, calculations | Medium |
| UI behavior | components with state/handlers | Medium |
| Config/infra | env handling, CI, feature flags, dependencies | Medium (High if secrets/deploy) |
| Presentation only | styles, copy, static assets | Low |
| Tests/docs | test files, markdown | Low (but note DELETED tests as High) |
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.
- 12d ago First seen · 190 lines · 54 tokens per session scan A f6f6d0d83c6b
AI Release Guardian is a skill published in the GitHub repository PramodDutta/qaskills (223 stars, last pushed 13d ago), licensed MIT. It adds 54 tokens to every session and 2,430 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
release-verification
A release-verification checklist for software versions and packages. It covers changelogs, tests, packaging, installation smoke checks, version-control actions, publishing, and checks after release.
release-testing-workflow
Use this skill when you need release-phase QA workflow from T-14 planning to go/no-go and post-release monitoring; triggers include release testing workflow and go/no-go QA.
orchestrate-release-engineer
Run the final release gate: fresh-clone simulation, artifact freshness, packaging hygiene, documentation accuracy. Use in the final hour. Catches the defects that make a working submission fail on someone else's machine.
pack-test
Test the claude-second-brain npm package by packing it into a tarball and running it via npx. Exercises the full multi-brain scaffold flow end-to-end: scaffolds a brain, verifies config.toml, global skills, and local patches, then scaffolds a second brain to validate config upsert. Trigger phrases: /pack-test, test…
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
developer-release
Release management, changeset workflow, firewall log parsing, breaking CLI change rules, and Go module summaries for gh-aw.