Borrowing it
Nothing to install: this file belongs to vlad-ryzhkov/ai-context-engineering-for-qa. 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/vlad-ryzhkov/ai-context-engineering-for-qa/main/.claude/skills/output-review/SKILL.mdgit clone --depth 1 https://github.com/vlad-ryzhkov/ai-context-engineering-for-qaWrote 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/vlad-ryzhkov/ai-context-engineering-for-qa/output-review)<a href="https://agentmods.dev/skills/vlad-ryzhkov/ai-context-engineering-for-qa/output-review"><img src="https://agentmods.dev/badge/skills/vlad-ryzhkov/ai-context-engineering-for-qa/output-review/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/vlad-ryzhkov/ai-context-engineering-for-qa/output-review"><img src="https://agentmods.dev/badge/skills/vlad-ryzhkov/ai-context-engineering-for-qa/output-review.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 | $0.00042 | $0.02974 |
| Opus 5 | $0.00021 | $0.01487 |
| Sonnet 5 | $0.00008 | $0.00595 |
| Haiku 4.5 | $0.00004 | $0.00297 |
Grade A, and why
output-review 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 2d 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 — 327 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/output-review — Independent Audit of Skill Output
Verifies skill OUTPUT against checklists from the target skill's SKILL.md. Independent assessment — not the same AI context that generated the result.
Before Starting
Read .claude/qa_agent.md.
When to Use
- Immediately after any skill completes (
SKILL COMPLETE) - When Score in SKILL COMPLETE seems inflated
- For independent validation before merge/release
Input Data
| Parameter | Required | Description |
|---|---|---|
| skill-name | Optional | Skill name (/output-review api-tests). If not specified — searches for SKILL COMPLETE in chat |
Algorithm (7 Phases)
Verbosity Protocol
Structured Output Priority: All analysis goes into the artifact (MD/HTML), not into chat.
Chat output (constraints):
- Brief Summary: max 5 lines (what was found, how many, result)
- Findings table: max 15 lines (top by severity)
- Full report:
📊 Full report: {path}+ open file
Iterative steps: Do not output progress for each file. Checkpoint only when:
- Phase transition (Phase N → Phase N+1)
- Blocker detected
- Completion (SKILL COMPLETE)
Tools first:
- Grep → table → report, without "Now I will grep..."
- Read → analyze → report, without "The file shows..."
Post-Check: Inline before SKILL COMPLETE (5-7 line checklist), not a separate file.
Phases 1-6: Silent. Phase 7: Save full report to file + brief summary in chat (max 5 lines).
Phase 1 — Target Identification
Goal: Determine which skill to audit.
- If parameter
/output-review {skill-name}is specified → use it - Otherwise — search for the last
SKILL COMPLETE: /{skill-name}in chat context - Fallback — ask the user: "Which skill to audit?"
- Validation: Glob
.claude/skills/{skill-name}/SKILL.md— file MUST exist
If skill not found → STOP:
❌ Skill /{skill-name} not found in .claude/skills/
Available: [list from Glob]
Phase 2 — Checklist Extraction
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.
- 2d ago Changed · +307 lines · +29 tokens per session f2cea3df7d4b
- 11d ago First seen · 20 lines · 13 tokens per session scan A 2ed1881228eb
output-review is a skill published in the GitHub repository vlad-ryzhkov/ai-context-engineering-for-qa (6 stars, last pushed 2d ago), licensed Unlicense. It adds 42 tokens to every session and 2,974 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 2026-08-31.
Other skills, from other repositories
detecting-broken-object-property-level-authorization
Detect and test for OWASP API3:2023 Broken Object Property Level Authorization vulnerabilities including excessive data exposure and mass assignment attacks.
Phone Use
MUST Load this skill for ANY task that involves GUI interaction or device control — opening apps, tapping, typing, scrolling, swiping, Back/Home/Recents, reading the screen, or any workflow that touches on-screen content. Load this even when the user does not explicitly mention the device — if the task implies a GUI…
release-new-version
Use when the user wants to release a new Zafiro version — drafting bilingual release notes, deciding the next version number, bumping app/build.gradle.kts, tagging, and publishing to GitHub (main repo, optionally the Xposed repo).
audit-brand
Audits and evolves brand positioning, voice consistency, and messaging for an existing brand. Works across personal brands and DTC/ecommerce store brands. Runs a structured audit across positioning, voice, visual direction, and content pillars, then produces a scorecard with specific fix recommendations. Also develops…
compliance-check
Assesses US regulatory compliance likelihood for any consumer product before listing for US sale. Accepts a CJ product PID/URL, Amazon ASIN, or product name/description as input. Classifies the product into a regulatory category (children's toys, electronics, cosmetics, supplements, pet products, apparel…
stress-test
Three-phase strategic decision analysis combining verbalized sampling (Stanford arXiv 2510.01171), customizable analytical lenses, and a structured decision brief. Surfaces tail-distribution insights — the non-obvious, suppressed analyses that mode-collapsed prompting misses. Adapted from @alexprompter's Model Council…