review-agentready

A command for reviewing a pull request against AgentReady's project-specific code-quality criteria. It maps findings to 25 attributes and estimates how they affect the project's self-assessment score.

In plain words
What is it for?
Use it to review an eligible pull request, identify compliance and quality problems, assign severity and confidence, and produce suggested remediation commands.
Why use it?
It combines normal code review with checks for project instructions, formatting, tests, and other AgentReady requirements, so reviewers can see both the issue and its likely score impact.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/ambient-code/agentready/review-agentready
Clone the repo
git clone --depth 1 https://github.com/ambient-code/agentready

Made for: Claude Code.

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,623 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00012 $0.01623
Opus 5 $0.00006 $0.00812
Sonnet 5 $0.00002 $0.00325
Haiku 4.5 $0.00001 $0.00162

Measured 2d ago against content hash da7f500e3663, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-agentready scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/review-agentready.md · 193 lines

How it starts

The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Provide an AgentReady-specific code review for the given pull request.

This command extends the standard /code-review with agentready project-specific concerns:

  • Map findings to the 25 agentready attributes (from agent-ready-codebase-attributes.md)
  • Calculate impact on self-assessment score (current: 80.0/100 Gold)
  • Generate remediation commands (agentready bootstrap, black, pytest, etc.)
  • Link to relevant CLAUDE.md sections
  • Categorize by severity with confidence scoring

Process

Follow these steps precisely:

  1. Eligibility Check (Haiku agent)

    • Check if PR is closed, draft, or already reviewed
    • Skip if automated PR (dependabot, renovate) unless it touches assessors
    • If ineligible, exit early
  2. Context Gathering (Haiku agent)

    • List all CLAUDE.md files (root + modified directories)
    • Get full PR diff and metadata
    • Return concise summary of changes
  3. Parallel AgentReady-Focused Review (5 Sonnet agents) Launch 5 parallel agents to independently review:

    Agent #1: CLAUDE.md Compliance Audit

    • Check adherence to CLAUDE.md development workflows
    • Verify pre-push linting (black, isort, ruff)
    • Verify test requirements (pytest, >80% coverage)
    • Check branch verification, conventional commits

    Agent #2: AgentReady-Specific Bug Scan Focus on agentready assessment logic:

    • TOCTOU bugs in file system operations
    • AST parsing correctness (false positives/negatives in assessors)
    • Measurement accuracy issues
    • Type annotation correctness
    • Error handling patterns (try-except, graceful degradation)

    Agent #3: Historical Context Analysis

    • Read git blame for modified assessor files
    • Check for regression in assessment accuracy
    • Verify attribute scoring logic hasn't changed unintentionally

    Agent #4: Previous PR Comment Analysis

    • Review comments on past PRs touching same files
    • Check for recurring issues

    Agent #5: Code Comment Compliance

    • Verify changes follow inline comment guidance
  4. Attribute Mapping (Haiku agent for each issue) For each issue found:

    • Map to specific agentready attribute ID (e.g., "2.3 Type Annotations")
    • Determine tier (1=Essential, 2=Critical, 3=Important, 4=Advanced)
    • Calculate score impact using tier weights (Tier 1: 50%, Tier 2: 30%, etc.)
    • Generate remediation command (black, pytest, agentready bootstrap --fix)
    • Link to CLAUDE.md section
  5. Confidence Scoring (parallel Haiku agents) For each issue, score 0-100 confidence:

    • 0: False positive
    • 25: Might be real, unverified
    • 50: Verified but minor
    • 75: Very likely real, important
    • 90: Critical issue (auto-fix candidate)
    • 100: Blocker (definitely auto-fix)

    Critical Issue Criteria (confidence ≥90):

    • Security vulnerabilities (path traversal, injection)
    • TOCTOU race conditions
    • Assessment accuracy bugs (false positives/negatives)
    • Type safety violations causing runtime errors
    • Missing error handling leading to crashes
  6. Filter Issues

    • Keep issues with confidence ≥80 for reporting
    • Flag issues with confidence ≥90 as "auto-fix candidates"
    • Calculate aggregate score impact
  7. Final Eligibility Check (Haiku agent)

    • Verify PR is still eligible for review
    • Check if PR was updated during review
  8. Post Review Comment (using gh pr comment) Use the custom AgentReady format (see below)

Read the full file on GitHub · 193 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 193 lines · 12 tokens per session scan A da7f500e3663

Subscribe to this mod's changes

review-agentready is a command published in the GitHub repository ambient-code/agentready (151 stars, last pushed 6d ago), licensed MIT. It adds 12 tokens to every session and 1,623 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.