Obliteratus
RedWoodOG/Hermes-Desktop
Remove refusal behaviors from open-weight LLMs using OBLITERATUS — mechanistic interpretability techniques (diff-in-means, SVD, whitened SVD, LEACE, SAE decomposition, etc.) to excise guardrails…
Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.
Install Implementation Final Review
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install openai/openai-guardrails-js implementation-final-review --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/implementation-final-review .claude/skills/implementation-final-review && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .claude/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-reviewType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install openai/openai-guardrails-js implementation-final-review --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/implementation-final-review .agents/skills/implementation-final-review && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .agents/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install openai/openai-guardrails-js implementation-final-review --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/implementation-final-review .cursor/skills/implementation-final-review && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .cursor/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/openai/openai-guardrails-js.git --path .agents/skills/implementation-final-review--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install openai/openai-guardrails-js implementation-final-review --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/implementation-final-review .gemini/skills/implementation-final-review && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .gemini/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install openai/openai-guardrails-js implementation-final-reviewInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/implementation-final-review .github/skills/implementation-final-review && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .github/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openai/openai-guardrails-js --skill implementation-final-review -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install openai/openai-guardrails-js implementation-final-review --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openai/openai-guardrails-js.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/implementation-final-review .opencode/skills/implementation-final-review && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "implementation-final-review" agent skill from https://github.com/openai/openai-guardrails-js/tree/main/.agents/skills/implementation-final-review into .opencode/skills/implementation-final-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "implementation-final-review", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
implementation-final-review- GitHub stars
- 105
- Token cost
- ~825 tokens
- SKILL.md length
- 415 words
- Files
- 3 (incl. references)
- Skills in repo
- 5
- Repo updated
- First seen
- Licence
- MIT
At a glance
Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.
- Tasks that involve LLM guardrails
- SKILL.md covers Prepare the complete scope, Dispatch and converge and Close the review
- Calls git
What it does
Implementation Final Review is an agent skill from openai/openai-guardrails-js, published by the product's own GitHub organization. Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.
Its SKILL.md is about 830 tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/reviewer-brief.md`).
It sits in AI & LLM Engineering, covering LLM guardrails. The repository describes itself as: OpenAI Guardrails - TypeScript / JavaScript. The licence is MIT.
When your agent uses it
- Tasks that involve LLM guardrails
Example prompts
- “/implementation-final-review”
What it can do on your machine
Read from SKILL.md and the folder at commit d28dc28. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Implementation Final Review loads about 825 tokens when it runs, and up to ~1.4k if it reads all its reference files. Until then it costs about 43 tokens; SKILL.md has 415 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from openai/openai-guardrails-js at commit d28dc28, republished under its MIT licence (© openai). 415 words, ~825 tokens.
Download SKILL.mdSave it as
.claude/skills/implementation-final-review/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.- name
- implementation-final-review
- description
- Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.
Implementation Final Review
Apply the adversarial-review procedure in AGENTS.md.
Use the installed $adversarial-review skill when available; otherwise follow
that inline procedure. This skill supplies a reusable brief, not another gate.
It applies to repository workflow changes as well as runtime changes before a
push or PR update.
Prepare the complete scope
Record the exact selected linked worktree, current HEAD, intended target, and comparison base SHA. Resolve the merge base with that target; for an explicitly requested stack, the previous PR's branch is the target. Inspect both the full stack and each incremental PR when reviewing several stacked changes together. Reviewers must check that every intermediate PR stands alone and references only files or skills present at that point.
Include all branch commits from the comparison base, staged and unstaged
changes, and relevant untracked deliverables. Ordinary git diff omits
untracked files; provide their paths and contents explicitly. Distinguish local
working notes from shipped files. Include individual commit diffs so an endpoint
diff cannot conceal unrelated changes. Freeze task content while reviewers run.
Read reviewer-brief.md when preparing dispatch. Fill it with the original outcome and constraints; do not include previous reviewers' conclusions or the implementer's preferred answer.
Show full SKILL.md (220 more words)Show less
Dispatch and converge
For each round, spawn exactly two new independent, read-only subagents with
fork_turns="none", both in this same worktree. Spawn both before waiting.
Give each a complete self-contained brief. Reviewer A checks correctness,
compatibility, security boundaries, and evidence. Reviewer B checks ownership,
architecture, maintainability, and unnecessary complexity. Both inspect the
entire change. They must not edit, modify Git state, or spawn agents.
Aggregate supported findings and classify them using AGENTS.md. Fix only in-scope blockers. Require two consecutive clean rounds on unchanged content, with a fresh pair in every round and no more than ten rounds. A substantive change resets the clean-round count. Do not substitute a self-review or reused agent when fresh reviewers are unavailable; stop before pushing and report the limitation. Do not create additional tasks or worktrees for reviewers.
Close the review
Run applicable verification, including focused evidence for repairs. Review any touched security surface and check Changesets requirements. Internal-only workflow changes need no package release note. Preserve commit hooks; if committing or later edits change reviewed content, repeat the invalidated checks and review.
Report scope, reviewer rounds, resolved in-scope findings, separate follow-ups, and actual verification results. Record evidence in working notes outside the shipped diff; no review-state certificate or component-credit reuse is needed. A clean local review does not authorize a merge or replace current-head CI.
© openai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 2 other files (references) in .agents/skills/implementation-final-review of openai/openai-guardrails-js.
- SKILL.md
- agents/openai.yaml
- references/reviewer-brief.md
Open the folder on GitHubat commit d28dc28
Compare with similar skills
Implementation Final Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Implementation Final Review this skillopenai/openai-guardrails-js | 105 | — | ~825 | Automated safety check: Pass | MIT | |
| ObliteratusRedWoodOG/Hermes-Desktop | 177 | 6 repos | ~3.8k | Automated safety check: Pass | MIT | |
| Lemonade Router Builderamd/skills | 398 | — | ~4k | Automated safety check: Pass | MIT | |
| Wp Project Triagegambitph/Stackable | 350 | 4 repos | ~371 | Automated safety check: Pass | GPL-3.0 | |
| Aisafetyhotwuyoscar/AISafetyHot-Hub | 224 | — | ~774 | Automated safety check: Pass | Custom licence | |
| Persona Designkangarooking/system-prompt-skills | 205 | 1 repos | ~956 | Automated safety check: Pass | MIT |
Similar skills
Lemonade Router Builder
amd/skills
Turns a natural-language description of routing intent into a valid Lemonade collection.router policy JSON.
Wp Project Triage
gambitph/Stackable
A skill your agent uses when you need a deterministic inspection of a WordPress repository (plugin/theme/block theme/WP core/Gutenberg/full site) including tooling/tests/version hints, and a…
Aisafetyhot
wuyoscar/AISafetyHot-Hub
Read AI Safety HOT daily digests, search recent AI safety research and incidents, and follow current hot topics.
Persona Design
kangarooking/system-prompt-skills
当需要为 AI 产品定义核心身份、角色声明和能力边界时调用此 skill。典型场景包括：设计新 AI 产品的 system prompt 首段、为不同场景创建差异化角色（如教学助手 vs 编程代理）、重新定义 AI 与用户的关系框架。
Execution Guardrails
mrtooher/fable-mode
Always-on operational guardrails, model-independent. An agent skill from mrtooher/fable-mode.
More from openai/openai-guardrails-js
- Official
Code Change Verification
openai/openai-guardrails-js
Select and run Guardrails verification for code, dependency, test, tooling, documentation, or repository-workflow changes.
- Official
Implementation Kickoff
openai/openai-guardrails-js
Start or resume a requested Guardrails implementation or PR takeover in the selected linked worktree with bounded scope and verification.
- Official
Implementation Strategy
openai/openai-guardrails-js
Choose bounded implementation scope and existing owning modules before Guardrails runtime, configuration, client, resource, streaming, Agents, or SDK compatibility changes and feedback fixes.
- Official
PR Draft Summary
openai/openai-guardrails-js
Draft a Guardrails PR from its complete final diff and carry out authorized CI and review follow-up, including stacked PRs and takeovers.
Related
Categories
Questions about Implementation Final Review
What does Implementation Final Review do?
Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure. Implementation Final Review is an agent skill from openai/openai-guardrails-js, published by the product's own GitHub organization. Prepare independent review of a complete Guardrails change before pushing or updating a PR, using the repository adversarial-review procedure.
When should I use Implementation Final Review?
Implementation Final Review fits situations like: tasks that involve LLM guardrails.
How do I install Implementation Final Review in Claude Code?
Run `npx skills add openai/openai-guardrails-js --skill implementation-final-review -a claude-code`. Or copy the skill folder (.agents/skills/implementation-final-review in openai/openai-guardrails-js) into .claude/skills/implementation-final-review in your project. Claude Code loads it when a task matches its description.
How do I install Implementation Final Review in Codex?
Run `npx skills add openai/openai-guardrails-js --skill implementation-final-review -a codex`. Or copy the skill folder (.agents/skills/implementation-final-review in openai/openai-guardrails-js) into .agents/skills/implementation-final-review in your project. Codex loads it when a task matches its description.
Can I use Implementation Final Review in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openai/openai-guardrails-js --skill implementation-final-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/implementation-final-review, .gemini/skills/implementation-final-review, .github/skills/implementation-final-review and .opencode/skills/implementation-final-review in your project.
What does Implementation Final Review need to run?
Going by SKILL.md and its folder, Implementation Final Review needs the command-line tools its instructions call (git).
Does Implementation Final Review access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Implementation Final Review safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Implementation Final Review use?
Implementation Final Review is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Implementation Final Review use?
About 825 tokens (SKILL.md is roughly 3.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 581 tokens, read only when the agent opens those files.
What are the alternatives to Implementation Final Review?
Skills that share tags, products or a category with Implementation Final Review: Obliteratus (RedWoodOG/Hermes-Desktop, 177 stars), Lemonade Router Builder (amd/skills, 398 stars), Wp Project Triage (gambitph/Stackable, 350 stars) and Aisafetyhot (wuyoscar/AISafetyHot-Hub, 224 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Implementation Final Review?
openai (a GitHub organization, an official publisher) maintains it in openai/openai-guardrails-js, which has 105 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 8, 2026.
Source: openai/openai-guardrails-js on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.