llm-output-reviewer SKILL.md skillA
llm-output-reviewer from polakinio/ai-engineering-playbook is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add polakinio/ai-engineering-playbook/llm-output-reviewer@git:20260812.5b0dd88Pins git:20260812.5b0dd88 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- polakinio/ai-engineering-playbook
- Path
skills/llm-output-reviewer/SKILL.md- Versions recorded
- 3
- Latest version
- git:20260812.5b0dd88, committed Aug 12, 2026
- First version recorded
- Mar 23, 2026
- Size
- 2,201 bytes
- Repository stars
- 2
- License
- MIT
Review or plan changes that affect prompts, model output validation, fallback behavior, or LLM-driven decisions.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# llm-output-reviewer ## Purpose During Development Use this skill when Codex and similar AI coding agents are reviewing or implementing prompt logic, model output validation, fallback behavior, model-driven decisions, or other LLM-integrated code paths. This is a reusable development-time reasoning framework. It is not runtime code and must not be coupled into the product being reviewed. ## When To Apply - prompt or system-instruction changes - output-schema or validation changes - fallback-path changes for failed or missing model responses - PRs that make product behavior depend on model output ## What To Inspect - prompt construction - input shaping - output validation - fallback behavior - downstream consumers of model output - tests or fixtures covering malformed responses ## Rules To Enforce - raw model output must not be trusted directly - validated structure must match downstream expectations - fallback behavior must remain safe and operational - prompts must not silently demand a new output shape without validation changes - model-driven decisions must remain bounded by repository policy ## Common Anti-Patterns To Catch …
Read the whole file at its exact version.
How to install
mdr add polakinio/ai-engineering-playbook/llm-output-reviewer@sha256:1dcbd52b6d3c78eaThe line at the top of this page pins the label git:20260812.5b0dd88, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_f5neeyncbzpc4hsu)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Mar 23, 2026 and Aug 12, 2026, about one every 71 days. Its first recorded version is graded C and its latest A.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260812.5b0dd88 latest | 2026-08-12 | 5b0dd88 | 2,201 B | A | view · diff |
| git:20260329.e81eb98 | 2026-03-29 | e81eb98 | 1,968 B | C | view · diff |
| git:20260323.057cdcd | 2026-03-23 | 057cdcd | 1,951 B | C | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (2201 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
polakinio/ai-engineering-playbook · 2 stars · license MIT · pushed 2026-09-04 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_f5neeyncbzpc4hsu GET https://markdownregistry.com/api/v1/resolve?ref=polakinio/ai-engineering-playbook/llm-output-reviewer GET https://markdownregistry.com/api/v1/blob/1dcbd52b6d3c78ea6824f6fd4a2c2a8458f90a23970875f3064e21f48891e24f
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.