Skip to content
Agentmart
Submit a tool

multi-review

Perform a comprehensive code review using multiple specialized sub-agents in parallel.

Skill by phobologic

IndexedMIT★ 7
Write a review

Chalk Labs

Need help deploying multi-review?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

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/phobologic-multi-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.

We use your email only for this newsletter. Privacy

About

Perform a comprehensive code review using multiple specialized sub-agents in parallel. Covers logical correctness, performance, readability, and security. Creates tk tickets if tk is available, otherwise writes a report to .code-review/final-report.md.

Use cases: documents, security

No 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/multi-review

What's inside

Skill name
multi-review
Description
Perform a comprehensive code review using multiple specialized sub-agents in parallel. Covers logical correctness, performance, readability, and security. Creates tk tickets if tk is available, otherwise writes a report to .code-review/final-report.md.
License
MIT
Instructions
174 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Multiple Code Reviewers

This command performs a comprehensive code review using specialized sub-agents.

## Step 1: Setup and detect tk

Run these commands sequentially in a **single Bash call** to prepare the environment and detect tk:
```bash
mkdir -p .code-review
rm -f .code-review/*.md
touch .code-review/changed-files.txt
TK_AVAILABLE=false
if command -v tk >/dev/null 2>&1; then
  TK_AVAILABLE=true
fi
echo "TK_AVAILABLE=$TK_AVAILABLE"
```

## Step 2: Determine review scope

Parse `$ARGUMENTS` to determine what to review. Set two variables: the file list in `.code-review/changed-files.txt` and `REVIEW_CMD` (passed to reviewers).

- **No arguments** (default): Review uncommitted changes
  ```bash
  git status --porcelain | awk '{print $NF}' > .code-review/changed-files.txt
  ```
  `REVIEW_CMD=git diff`

- **Commit references** (e.g., "last 3 commits", "since abc123", "HEAD~5"):
  Determine the appropriate base ref from the user's intent, then:
  ```bash
  git diff --name-only <base-ref> > .code-review/changed-files.txt
  ```
  `REVIEW_CMD=git diff <base-ref> --`

- **Specific files** (e.g., "Review src/auth.py src/models.py"):
  Write the specified file paths to `.code-review/changed-files.txt` (one per line).
  `REVIEW_CMD=FULL_FILE`

It's CRUCIAL that reviewers ONLY analyze files listed in .code-review/changed-files.txt and NOTHING ELSE.

## Step 3: Collect CLAUDE.md context

Collect CLAUDE.md files from the project root and from directories containing changed files. This content is passed to reviewers so they can check for project convention violations.

Run this to build `.code-review/claude-md-context.txt`:
```bash
> .code-review/claude-md-context.txt
[ -f CLAUDE.md ] && { echo "=== CLAUDE.md (root) ==="; cat CLAUDE.md; echo; } >> .code-review/claude-md-context.txt
while IFS= read -r file; do
  dir=$(dirname "$file")
  if [ "$dir" != "." ] && [ -f "$dir/CLAUDE.md" ]; then
    echo "=== CLAUDE.md ($dir) ===" >> .code-review/claude-md-context.txt
    cat "$dir/CLAUDE.md" >> .code-review/claude-md-context.txt
    echo >> .code-review/claude-md-context.txt
  fi
done < .code-review/changed-files.txt
```

If the resulting file is empty (no CLAUDE.md files found), delete it so reviewers know there is no context:
```bash
[ ! -s .code-review/claude-md-context.txt ] && rm -f .code-review/claude-md-context.txt
```

## Step 4: tk mode setup (if TK_AVAILABLE)

If `TK_AVAILABLE` is true:

1. Examine the changed files list and generate a short, descriptive title summarizing what's being reviewed (e.g., "Review: Auth module refactor" or "Review: API endpoint updates and test fixes").
2. Create an epic with a timestamp for uniqueness:
```bash
EPIC_ID=$(tk create "<generated title> (<YYYY-MM-DD HH:MM>)" -t epic -p 2 --tags code-review -d "<brief summary of changes being reviewed>")
```
3. Note the `EPIC_ID` for passing to sub-agents.

## Step 5: Launch reviewers

If there are files to review, use the Task tool to invoke these specialized reviewers in parallel.

Include `REVIEW_CMD=<command>` in each prompt so reviewers know how to examine files.

### tk mode (TK_AVAILABLE=true)

Pass the epic ID and review command to each sub-agent via the prompt string:

1. Invoke the code-reviewer-1 sub-agent:
```
Task: Review code for logical correctness
Prompt: TK_MODE=true EPIC_ID=<epic_id> REVIEW_CMD=<review_cmd> -- Review ONLY files in .code-review/changed-files.txt - do NOT examine any other files
SubagentType: code-reviewer-1
```

2. Invoke the code-reviewer-2 sub-agent:
```
Task: Review code for performance
Prompt: TK_MODE=true EPIC_ID=<epic_id> REVIEW_CMD=<review_cmd> -- Review ONLY files in .code-review/changed-files.txt - do NOT examine any other files
SubagentType: code-reviewer-2
```

3. Invoke the code-reviewer-3 sub-agent:
```
Task: Review code for readability
Prompt: TK_MODE=true EPIC_ID=<epic_id> REVIEW_CMD=<review_cmd> -- Review ONLY files in .code-review/changed-files.txt - do NOT examine any other files
SubagentType: code-reviewer-3
```

4. Invoke the security-reviewer sub-agent:
```
Task: Review code for security issues
Prompt: TK_MODE=true EPIC_ID=<epic_id> REVIEW_CMD=<review_cmd> -- Review ONLY files in .code-review/changed-files.txt - do NOT examine any other files
SubagentType: security-reviewer
```

### File mode (TK_AVAILABLE=false)

Pass the review command without tk parameters:

1. Invoke the code-reviewer-1 sub-agent:
```

From phobologic/claude_code, MIT.

Signals

Stars
7
Forks
0
Open issues
0
Last commit (this folder)
Mar 6, 2026 (7 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from phobologic

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get multi-review running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying multi-review?

CallGet help
Cite this page

multi-review skill. Agent Mart. https://agentmart.dev/skills/phobologic-multi-review (last updated 2026-10-24)