agentleFS
codewithmukeshcodewithmukesh/dotnet-claude-kit/skills/code-review/SKILL.md

code-review

Skill808 starsChanged 4 months agoMITView on GitHub

MCP-powered multi-dimensional code review for .NET projects. Uses Roslyn analysis tools for antipatterns, diagnostics, references, and dependency graphs combined with structured manual review. Prioritizes effort with blast-radius scoring — data access, security, concurrency, and integration boundaries before style — and produces severity-categorized findings with actionable fixes. Use when: "review", "code review", "PR review", "review this", "review my code", "check code quality", "review changes", "what should I review", "review priorities", "blast radius", "critical path".

194 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use close-spec, design-with-images and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • dzhngSkill

    close-spec

    When a spec under specs/<feature>/ is done shipping (a write-spec build, or any planned task that produced a spec), archive it to specs/done/ and rewrite it from a build-plan into a durable rationale — the why, the principles, the invariants — pointing back to the real code for the how. Use when implementation has landed and the plan no longer matches what shipped, or the user says a feature/spec is finished. Pairs with [write-spec](../write-spec/SKILL.md) (the plan this closes) and [review](../review/SKILL.md) (run before closing).

    Also in 5 projects that use this

  • dzhngSkill

    design-with-images

    Explore UI and visual design with image generation, then iterate the real implementation against the user-selected concept. Use when brainstorming visual directions, offering mockup variants, or implementing an approved generated design; do not stop at the mockups or the first code approximation.

    Also in 5 projects that use this

  • PydanticSkill

    add-new-model

    Add support for a newly-released language or image generation model in pydantic-ai (e.g. openai:gpt-5.6, anthropic:claude-sonnet-5, openai:gpt-image-2). Use when a provider ships a new model id and you need to wire literals, profile flags, adapters, and tests to recognize it. Handles SDK-lag, gateway list conventions, capability probing, and direct image-model geometry.

    Also in 9 projects that use this

  • transitive-bullshitSkill

    compare-screenshots

    Compare screenshots against the intended design, distinguishing approved references from historical baselines. Use for iterative implementation matching, before/after visual review, objective image telemetry, or checking a lone capture for flat, empty, or badly framed content.

    Also in 5 projects that use this

  • Microsoft AzureSkill

    azsdk-common-apiview-feedback-resolution

    Analyze and resolve APIView review feedback on Azure SDK PRs. **UTILITY SKILL**. USE FOR: APIView comments, API review feedback, SDK API surface changes. DO NOT USE FOR: general code review, non-APIView feedback. INVOKES: azure-sdk-mcp:azsdk_apiview_get_comments, azure-sdk-mcp:azsdk_customized_code_update, azure-sdk-mcp:azsdk_typespec_delegate_apiview_feedback, azure-sdk-mcp:azsdk_run_typespec_validation, azure-sdk-mcp:azsdk_package_generate_code.

    Also in 6 projects that use this

  • Microsoft AzureSkill

    azsdk-common-generate-sdk-locally

    Generate and finalize Azure SDKs locally from TypeSpec, including build, checks, tests, changelog, metadata, and version updates by default. WHEN: "generate SDK locally", "build SDK", "run SDK tests", "run CI checks", "validate package", "run checks", "update changelog", "fix SDK build errors", "resolve SDK generation errors", "customize TypeSpec", "rename SDK client", "rename SDK model", "hide operation from SDK", "fix analyzer errors", "resolve customization drift", "create subclient", "update metadata", "update version". INVOKES: azsdk_verify_setup, azsdk_package_generate_code, azsdk_package_build_code, azsdk_package_run_check, azsdk_package_run_tests, azsdk_customized_code_update, azsdk_package_update_changelog_content, azsdk_package_update_metadata, azsdk_package_update_version.

    Also in 6 projects that use this

  • +4 more from dzhng
  • +2 more from Azure

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 184 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After a major refactor to verify no regressions or design drift

    line 32
  2. Distinctive:

    Check whether callers handle changed return types and new error cases.

    line 69
  3. Very specific:

    Verify dependency direction (Domain → nothing.

    line 73
  4. Distinctive:

    Every finding states what's wrong, why it matters, and how to fix it.

    line 96

“Common” is measured: kept by at least 6 in 10 of the 184 edited versions of this file we could read. “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

/code-review — MCP-Powered Code Review

What

Performs a multi-dimensional code review combining Roslyn MCP analysis with structured manual review. Effort follows the 80/20 rule: the 20% of code that causes 80% of incidents (data access, security, concurrency, integration boundaries) gets thorough review; style and formatting are left to tooling.

Review dimensions: Correctness (logic, edge cases, null handling, async pitfalls), Security (auth gaps, injection, secrets, CORS), Performance (N+1, allocations, missing cancellation), Architecture compliance (layer violations, boundary breaches), Test coverage (behavior tests for changed types).

When

  • "Review this", "code review", "PR review", before merging a pull request
  • After a major refactor to verify no regressions or design drift
  • "What should I review?" — deciding where review effort goes on a large change
  • Onboarding to unfamiliar code and wanting a quality assessment

How

Step 1: Scope and Score Blast Radius

Identify changed files (git diff main...HEAD, specified files, or module). Score each change to set review depth — blast radius determines depth, not line count. A one-line middleware change outranks a 300-line rename.

Blast RadiusExamplesDepth
CriticalMiddleware, auth, DB migrations, shared kernel, CI/CDThorough — every code path
HighPublic API changes, message consumers, EF configuration, new moduleFocused — consumers + behavior
MediumNew feature following existing patterns, bug fix, new endpointStandard — checklist pass
LowDocs, formatting, renames, logging statementsGlance — build + tests pass

Step 2: MCP Analysis (before reading any file)

detect_antipatterns(projectFilter: "affected-project")   → async void, DateTime.Now, new HttpClient(), broad catch
get_diagnostics(scope: "project", path: "affected-project") → new warnings, nullability issues

Distinguish newly introduced findings from pre-existing ones — focus on new.

Step 3: Blast Radius Verification

For each modified public API:

find_references(symbolName: "ModifiedType")              → count consumers; high count = high risk
get_dependency_graph(symbolName: "ModifiedMethod", depth: 2) → ripple effects

Check whether callers handle changed return types and new error cases.

Step 4: Architecture Compliance

Verify dependency direction (Domain → nothing; Infrastructure → Application → Domain) via get_project_graph and detect_circular_dependencies. Per architecture: VSA features don't cross-reference; Clean Architecture domain has zero project references; Modular Monolith modules communicate only via integration events — find_references on a module's DbContext should resolve only inside that module.

Step 5: Manual Review — Priority Order

Review what tools can't catch, highest-risk areas first:

PriorityAreaCheck
1Data accessN+1 (missing Include/projection), raw SQL with user input, missing CancellationToken
2SecurityEvery endpoint has explicit [Authorize]/[AllowAnonymous], input validated, no secrets in code, no PII in logs
3ConcurrencyToken propagated end-to-end, no .Result/.Wait(), thread-safe shared state
4IntegrationRetry/timeout on external calls, consumer idempotency, no swallowed exceptions
5CorrectnessBusiness logic, edge cases (empty/null/concurrent), entities mapped to DTOs at the boundary
6TestsBehavior tested (not implementation); happy path + main error case covered
—Style/namingMention only after the above; formatters and analyzers own this

Step 6: Produce the Review

Every finding states what's wrong, why it matters, and how to fix it. Never bury a security bug under naming nits.

## Code Review: [Scope]

### Summary
[1-3 sentences: scope, risk level, recommendation]

### Critical (must fix before merge)
- **[Title]** — [file:line] [What's wrong. Why it matters. How to fix.]

### Warnings (should fix, creates tech debt)
- **[Title]** — [file:line] [...]

### Suggestions (nice to have)
- **[Title]** — [file:line] [...]

### Architecture Compliance
[PASS/WARN with boundary-violation notes]

### Test Coverage
[Which changed types have tests; specific scenarios to add]

### What's Good
- [Always include — reinforce good patterns]

Quick review (1-2 files, low blast radius): run detect_antipatterns + get_diagnostics, read for correctness, output Summary + Issues + What's Good.

Example

User: /code-review the changes in this PR

Claude: 7 changed files across 3 projects. CreateOrder touches data access
and a public endpoint — High blast radius. Running MCP analysis...

## Code Review: Order Processing Feature

### Summary
Adds CreateOrder/GetOrder endpoints with EF Core persistence. Well-structured
VSA feature. Two issues need attention before merge.

### Critical (must fix before merge)
- **Missing CancellationToken propagation** — CreateOrder.cs:38
  SaveChangesAsync() called without the token. Client disconnects keep
  burning server resources. Pass `ct` from the handler parameter.

### Warnings (should fix, creates tech debt)
- **N+1 query in GetOrder** — GetOrder.cs:25
  Order loaded without `.Include(o => o.Items)`; one lazy load per item
  during serialization. Eager-load or use a projection.

### Suggestions (nice to have)
- **Seal the handler** — CreateOrderHandler.cs:10
  Not designed for inheritance; `sealed` enables devirtualization.

### Architecture Compliance
PASS — all changes within Features/Orders/, no layer violations.

### Test Coverage
Happy path covered. Add tests for validation failure and not-found.

### What's Good
- Clean command/query separation; FluentValidation covers edge cases
- Response DTOs are records, no entity leaks

Related

  • /de-sloppify — Cleanup pass for the style/formatting issues review skips
  • /verify — Automated verification pipeline (complements manual review)
  • /health-check — Broader project health assessment beyond a single PR

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.