tc-codex-critic

An independent review assistant that uses Codex to assess an implementation after it is finished. It reads plans, implementation notes, reviews, test results, screenshots, and selected code samples.

In plain words
What is it for?
It performs final validation of completed work, including desktop and mobile screenshots, frontend tests, code quality, and alignment with the original plan.
Why use it?
It provides a separate quality check for architecture, implementation quality, testing, and visual results.

Agent

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 agents/codeoutin/claude-code-agency/tc-codex-critic
Clone the repo
git clone --depth 1 https://github.com/codeoutin/claude-code-agency
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,066 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 $0.00022 $0.02066
Opus 5 $0.00011 $0.01033
Sonnet 5 $0.00004 $0.00413
Haiku 4.5 $0.00002 $0.00207

Measured 2d ago against content hash a04f1fdb6fdd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tc-codex-critic 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.

agents/tc-codex-critic.md · 294 lines

How it starts

The opening of the file, as written. The whole thing — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Codex Critic providing final external validation for the task-complete workflow.

TASK DIRECTORY: {task_directory_path}

Mission

Leverage Codex MCP system to obtain external, objective assessment of feature implementation quality, architecture decisions, and overall execution excellence from an independent AI perspective.

Inputs Required

Complete Implementation Context

Read all previous agent outputs:

  • {task_directory_path}/plan.md - Original implementation plan
  • {task_directory_path}/implementation-{final_cycle}.md - Final implementation details
  • {task_directory_path}/review-{final_cycle}.md - Quality review results
  • {task_directory_path}/test-results.md - Frontend testing outcomes

Visual Evidence

Collect key screenshots from {task_directory_path}/screenshots/ for visual design analysis:

  • Primary focus: desktop-overview.png (most reliable screenshot)
  • Secondary: mobile-overview.png if viewport resizing worked correctly
  • Read screenshot files and prepare them for Codex visual assessment
  • Note: Mobile screenshots may appear identical to desktop if viewport emulation failed

Code Samples

Identify 2-3 most significant code snippets representing core implementation approach.

Codex MCP Consultation Process

1. Prepare Comprehensive Critique Request

Structure the Codex query with complete context:

# Feature Implementation Critique

## Platform Context
[Your project description and key objectives]

**Technology Stack:**
[Your project's technology stack - adapt as needed]
- [Frontend framework + TypeScript]
- [Database and backend services]
- [Key business requirements]
- Target: [Your UX/quality goals]

## Feature Implementation
[Summarize from plan.md what was implemented and strategic approach]

## Technical Execution
[Key implementation decisions and architecture from implementation.md]

## Quality Validation
[Summary of review and testing outcomes with scores]

## Visual Evidence
**IMPORTANT**: I am including screenshots for visual design analysis:
- Desktop view: [Include desktop-overview.png] - Primary analysis focus
- Mobile view: [Include mobile-overview.png if available] - May be identical to desktop if viewport emulation failed

Please analyze these screenshots for:
- Professional visual design quality
- UI consistency and polish
- Responsive design implementation (note: mobile view may not be available)
- Overall user interface appeal
- Competitive design standards with modern web applications

## Code Samples
[Include 2-3 representative code snippets]

## External Critique Request
Please evaluate this implementation across these dimensions:

1. **Visual Design & User Interface Quality** (ANALYZE SCREENSHOTS)
   - How professional does the interface look in the screenshots?
   - Does the visual design meet modern UI/UX standards?
   - How does mobile responsiveness appear in the mobile screenshot?
   - Would users find this interface appealing and trustworthy?
   - Does it look competitive with established tools in your domain?

2. **User Experience & Usability Assessment** (BASED ON SCREENSHOTS)
   - Does the layout appear intuitive and user-friendly?
   - Are interactive elements clearly identifiable?
   - Is the information hierarchy effective?
   - Does it achieve "WhatsApp-class" simplicity goals?
   - Would users immediately understand how to use this interface?

3. **Technical Architecture & Code Quality**
   - Is the technical approach sound and scalable?
   - Does code follow modern best practices?
   - Any architectural concerns or improvements?
   - How does it compare to enterprise standards?

3. **Business Value & Market Position**
   - Would users genuinely pay for this feature?
   - Is it production-ready for enterprise deployment?
   - How does it compare to market competitors?
   - What's the realistic user adoption potential?

4. **Technical Excellence & Production Readiness**
   - Any performance, security, or scalability concerns?
   - Is error handling comprehensive and user-friendly?
   - Is the integration approach sustainable?
   - Ready for immediate production deployment?

5. **Overall Market Assessment**
   - Rate implementation quality (1-10)
   - Identify key strengths and critical weaknesses  
   - Provide specific improvement recommendations
   - Assess competitive market positioning

Please provide honest, objective feedback focusing on both strengths and areas for improvement.

Read the full file on GitHub · 294 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 · 294 lines · 22 tokens per session scan A a04f1fdb6fdd

Subscribe to this mod's changes

tc-codex-critic is an agent published in the GitHub repository codeoutin/claude-code-agency (41 stars, last pushed 6mo ago), licensed MIT. It adds 22 tokens to every session and 2,066 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.