Self Review logo

Self Review

Community
goatstarter
self-review

Quality review of your own diff before handing it over. Use after completing any multi-file or non-trivial change, before reporting it done. Catches AI-accent code (noise comments, debug leftovers, scope creep, orphan files) and diff-level bugs. Complements verify-done, which checks that the work RUNS; this checks that the code is CLEAN. Also invoked manually via /self-review.

Overview

Publishergoatstarter
Repositorygoat-fable
Skill nameself-review
Stars
62
Forks
8
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by goatstarter on GitHub. Read the source before you install it.

Installation

Install the Self Review AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/goatstarter/goat-fable/tree/main/skills/self-review
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/goatstarter/goat-fable.git /tmp/goat-fable
mkdir -p .claude/skills
cp -r /tmp/goat-fable/skills/self-review .claude/skills/self-review
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Self Review in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Self Review on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Self Review is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Self Review

Review your own diff as if a stranger wrote it and your name goes on the approval. Your context is biased toward what you meant to write, so this pass works from artifacts only: the actual diff and files, not your memory of editing them.

1. The full diff, top to bottom

Run git diff (and git status for untracked files). Read every hunk. You are hunting, in order of frequency:

  • Debug leftovers: console.log / print / dump statements added while working; temporary sleeps; verbose flags.
  • Scratch artifacts: test scripts, fixtures, output files you created to iterate and forgot. Untracked files you can't name a reason for: delete.
  • Commented-out code and TODOs you could resolve right now.
  • Hardcoded temporaries: paths, ports, ids, credentials, dates that were placeholders.
  • Orphans: imports/variables/functions your later edits made unused.
  • Accidents: deleted lines you don't remember deleting, files changed that the task had no business touching, formatting churn on untouched lines.

2. The AI-accent sweep

Would a reviewer smell generated code? Remove the accent:

  • Comments that narrate ("// fetch the user"), explain the change ("// updated to new API"), or tutor. Keep only constraint comments the code can't express.
  • Defensive bloat: null checks and try/catch on paths that cannot fail, validation of internal calls. Boundaries validate; internals trust.
  • Speculative generality: parameters nothing passes, options nothing sets, abstractions with one caller.
  • Convention mismatch: naming style, error idiom, file layout diverging from the neighboring code. The codebase's way wins over yours.

3. The requirements cross-check

Re-read the original request word by word. For each explicit and implicit requirement, point at the diff hunk that satisfies it. No hunk → it's not done, no matter how done it feels. This is where multi-part requests silently lose part 3 of 4.

Also the inverse: for each diff hunk, name the requirement it serves. Hunks serving no requirement are scope creep: revert them or justify them explicitly in the report.

4. The adversarial minute

Assume this diff contains exactly one bug and you must find it before the reviewer does. Best places to look: boundaries between changed and unchanged code, error paths you never exercised, the caller you updated from memory instead of grep, off-by-one at edges (empty list, first/last item, missing key).

For changes above ~150 lines or touching shared interfaces, don't trust your own adversarial minute: spawn the fresh-context verifier (agents/verifier.md pattern) and give it the requirements + diff, not your reasoning.

5. Close

Fix what you found (each fix re-enters this review), then proceed to verification (/verify-done) if you haven't run it. In the report, mention anything you noticed but deliberately left alone.

Frequently asked questions

What does the Self Review AI skill do?

Quality review of your own diff before handing it over. Use after completing any multi-file or non-trivial change, before reporting it done. Catches AI-accent code (noise comments, debug leftovers, scope creep, orphan files) and diff-level bugs. Complements verify-done, which checks that the work RUNS; this checks that the code is CLEAN. Also invoked manually via /self-review.

Why use Self Review on TypingMind?

Because you install it once and use it with any model. Self Review is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Self Review in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/goatstarter/goat-fable/tree/main/skills/self-review. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Self Review?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Self Review?

As many as you like. As long as a model supports skills, you can use Self Review with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Self Review AI skill free?

Yes. It is published on GitHub by goatstarter under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇