project-review
Review project files for consistency, completeness, correctness, and simplicity.
Skill by jmfontaine
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/jmfontaine-project-review
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, devops, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/project-review
What's inside
- Skill name
project-review- Description
- Review project files for consistency, completeness, correctness, and simplicity. Use when asked to review, audit, check, validate, or simplify project configuration files (justfile, Makefile, CI workflows, pre-commit config, pyproject.toml, Dockerfile, docker-compose, etc.) or documentation files (README, AGENTS.md, CLAUDE.md, CONTRIBUTING, etc.). Flags accidental complexity such as duplication, indirection, and dead or redundant settings. Supports both single-file review and cross-file checks. Not for source code, and not for checking the findings listed in a review report such as REVIEW.md.
- License
- Apache-2.0
- Instructions
- 102 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Project File Review
Review project files across four dimensions, then suggest fixes and improvements.
## Dimensions
- **Consistency**: Naming conventions, formatting, patterns, and style are uniform
within and across files.
- **Completeness**: Nothing is missing. All referenced targets, commands, variables, and
dependencies exist. All expected sections are present.
- **Correctness**: Syntax is valid. Commands work. Paths exist. Versions are compatible.
Logic is sound.
- **Simplicity**: No accidental complexity, meaning complexity the project's needs do
not require. Look for:
- Dead weight: unused targets, variables, dependencies, stages, hooks, and
documentation for removed features.
- Redundancy within the file: settings that restate tool defaults, duplicated steps or
blocks that could share a definition, and sections that repeat each other. Report
duplication across files in Phase 2 instead.
- Needless indirection: wrappers that only forward arguments, targets that only call
one other target, scripts that reimplement what a tool already does.
- Over-engineering: parameters, matrix entries, environment variables, or abstractions
with a single use or no use.
- Convoluted processes: multi-step instructions or workflows that a single command or
built-in tool feature could replace.
Every simplification must preserve required behavior. When a simplification drops
something (a platform, a version, a workflow), state what is lost so the user can
decide.
## Workflow
### Phase 1: Individual File Review
Review each file separately. For each file:
1. Read the file fully.
2. Evaluate against all four dimensions.
3. Report findings grouped by dimension.
4. Suggest specific fixes with code snippets.
### Phase 2: Cross-File Alignment
When multiple files are provided, or after individual reviews, check alignment across
files:
1. Verify shared references match (e.g., target names in justfile match CI workflow
steps, Python version in pyproject.toml matches CI matrix and Dockerfile).
2. Check that documented commands in README match actual targets/scripts.
3. Confirm dependency lists are synchronized (e.g., pyproject.toml vs requirements files
vs CI install steps).
4. Ensure agent instruction files (AGENTS.md, CLAUDE.md) reflect the actual project
structure and tooling.
5. Find duplication across files that could collapse to a single source of truth (e.g.,
a Python version repeated in four files instead of read from one, CI steps
reimplementing justfile recipes instead of calling them, CLAUDE.md copying AGENTS.md
instead of importing it). Skip overlap that serves different readers, such as README
and AGENTS.md both listing setup commands, unless one file can reference the other.
### Discovery Mode
When reviewing a file, suggest other project files that should be cross-checked. Common
relationships:
- **justfile / Makefile** <-> CI workflows, README, pre-commit config
- **pyproject.toml** <-> Dockerfile, CI workflows, pre-commit config
- **README** <-> AGENTS.md, CLAUDE.md, justfile, CI workflows
- **Dockerfile** <-> docker-compose, CI workflows, pyproject.toml
- **pre-commit config** <-> CI workflows, pyproject.toml
Offer to review related files the user hasn't mentioned yet.
## Output Format
For each file, report:
```text
### <filename>
#### Consistency
- [finding + suggested fix]
#### Completeness
- [finding + suggested fix]
#### Correctness
- [finding + suggested fix]
#### Simplicity
- [complexity + simpler alternative + anything lost]
```
After all individual reviews, add a cross-file section:
```text
### Cross-File Alignment
- [inconsistency or duplication + which files + suggested fix]
```
Omit empty sections. Prioritize actionable findings over nitpicks.From jmfontaine/agent-skills, Apache-2.0.
More from jmfontaine/agent-skills
- 0.570score
github-workflow
jmfontaine
Track work in GitHub issues and deliver it through pull requests and reviews.
SkillAFreshness: AClaude Code +9 - 0.570score
product-development-system
jmfontaine
Guide product development from manifesto and product map through releases, optional briefs, and release review.
SkillAFreshness: AClaude Code +9 - 0.570score
python-codebase-review
jmfontaine
Deep, whole-codebase review of a Python project that writes a new REVIEW.md, a verified and prioritized list of fixes and improvements.
SkillAFreshness: AClaude Code +9 - 0.566score
parallel-issues
jmfontaine
Work several GitHub issues in parallel with unattended coding agents, one Herdr workspace and git worktree per group of issues, each group delivered as a…
SkillAFreshness: AClaude Code +9 - 0.565score
git-commit
jmfontaine
Git commit guidelines, including which branch a commit goes on, the first commit, and GitHub stacked pull requests.
SkillAFreshness: AClaude Code +9 - 0.565score
review-verification
jmfontaine
Check every finding in an existing review or plan file (REVIEW.md or similar) against the current code or a named dependency version, mark each confirmed…
SkillAFreshness: AClaude Code +9 - 0.565score
git-branch
jmfontaine
Git branch naming guidelines, including the label to branch-type table.
SkillAFreshness: AClaude Code +9 - 0.558score
uv
jmfontaine
Python package management with uv.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 12, 2026 (17 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/jmfontaine/agent-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from jmfontaine
- 0.570score
github-workflow
jmfontaine
Track work in GitHub issues and deliver it through pull requests and reviews.
SkillAFreshness: AClaude Code +9 - 0.570score
product-development-system
jmfontaine
Guide product development from manifesto and product map through releases, optional briefs, and release review.
SkillAFreshness: AClaude Code +9 - 0.570score
python-codebase-review
jmfontaine
Deep, whole-codebase review of a Python project that writes a new REVIEW.md, a verified and prioritized list of fixes and improvements.
SkillAFreshness: AClaude Code +9 - 0.566score
parallel-issues
jmfontaine
Work several GitHub issues in parallel with unattended coding agents, one Herdr workspace and git worktree per group of issues, each group delivered as a…
SkillAFreshness: AClaude Code +9 - 0.565score
git-commit
jmfontaine
Git commit guidelines, including which branch a commit goes on, the first commit, and GitHub stacked pull requests.
SkillAFreshness: AClaude Code +9 - 0.565score
review-verification
jmfontaine
Check every finding in an existing review or plan file (REVIEW.md or similar) against the current code or a named dependency version, mark each confirmed…
SkillAFreshness: AClaude Code +9
Cite this page
project-review skill. Agent Mart. https://agentmart.dev/skills/jmfontaine-project-review (last updated 2026-10-24)