agentleFS
ivklgnivklgn/ai-kit/skills/review-golang/SKILL.md

review-golang

Skill12 starsChanged 2 months agoMITView on GitHub

Run a comprehensive Go code review on git-changed files using the golang-pro agent with Context7 library docs

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run these commands to find changed .go files:

    line 13
  2. Very specific:

    If NO changed .go files are found, use AskUserQuestion to ask the user which Go files or directories to review.

    line 22
  3. Very specific:

    Read go.mod to identify third-party dependencies (non-stdlib, non-internal modules).

    line 28
  4. Very specific:

    Use the Task tool with subagent_type: "golang-pro" to run the review.

    line 39

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Go Code Review Skill

You are performing a comprehensive Go code review. Follow these steps exactly.

Step 1: Identify Changed Go Files

Run these commands to find changed .go files:

git diff --name-only HEAD -- '*.go'
git diff --cached --name-only -- '*.go'

Combine and deduplicate the results. Exclude _test.go files from the primary review list but keep them for the test quality review area.

If NO changed .go files are found, use AskUserQuestion to ask the user which Go files or directories to review. Do not proceed without files.

Step 2: Pre-fetch Library Documentation via Context7 MCP

This step is critical — the golang-pro agent does NOT have Context7 MCP tools, so you must fetch docs now and pass them along.

  1. Read go.mod to identify third-party dependencies (non-stdlib, non-internal modules).
  2. Scan the changed Go files for import statements. Identify which third-party libraries are actually used.
  3. For each non-trivial third-party library found (skip stdlib, skip trivial/obvious ones): a. Call mcp__context7__resolve-library-id with the library's module path (e.g., github.com/spf13/cobra) b. If resolved, call mcp__context7__get-library-docs with the resolved ID and a topic query relevant to how the library is used in the changed code
  4. Collect the fetched documentation snippets — you will pass them to the golang-pro agent.

Limit to the 3-5 most important libraries to avoid overwhelming the review context.

Step 3: Launch golang-pro Agent for Review

Use the Task tool with subagent_type: "golang-pro" to run the review. Pass a prompt that includes:

  • The list of changed .go file paths (both source and test files)
  • The pre-fetched library documentation from Step 2
  • The review instructions below

Review Prompt Template

Construct the prompt as follows:

You are reviewing Go code changes. Read each file listed below and produce a structured review.

## Files to Review
{list of file paths}

## Library Documentation Context
{pre-fetched Context7 docs for third-party libraries used in these files}

## Review Areas

Analyze each file across these 8 areas:

1. **Go Idioms & Code Quality** — Non-idiomatic patterns, naming violations, struct design issues
2. **Error Handling** — Swallowed errors, missing error checks, incorrect error wrapping, sentinel vs type errors
3. **Memory & Allocations** — Unnecessary allocations, missing pre-allocation for known-size slices, string concatenation in loops. Only flag practical issues, no micro-optimizations.
4. **Performance Patterns** — Unoptimal algorithms, unnecessary work in hot paths, missing caching where obvious
5. **Dead Code** — Unused functions, unreachable branches, commented-out code left behind
6. **Code Duplication** — Repeated logic that would clearly benefit from extraction (3+ occurrences or complex blocks). Do NOT suggest extraction for simple 2-3 line patterns.
7. **Test Quality** — Missing edge cases, brittle assertions, test isolation issues, missing error case tests
8. **Library Usage** — Incorrect or suboptimal use of stdlib and third-party libraries. Use the library documentation provided above to verify correct API usage.

## Review Rules

- Be minimal and optimal — don't overcomplicate suggestions
- Be context-specific — no generic advice, only what matters for THIS code
- Simple > complex — never suggest an overengineered solution
- Every finding MUST include a file:line reference
- If a review area has no findings, omit it entirely
- Do NOT suggest adding comments, docstrings, or type annotations unless there is a clear correctness issue
- Do NOT suggest error handling for impossible scenarios
- Do NOT suggest abstractions for one-time code

## Output Format

Group findings by review area. For each finding:

**[Area Name]**
- `file.go:42` — Description of the issue and suggested fix

If everything looks good, say so briefly. Do not pad the review with praise or filler.

Step 4: Present Results

The golang-pro agent will return the structured review. Present it directly to the user. Do not add extra commentary or re-analyze — the review stands as-is.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.