Critic Agent — Development agent for Claude Code
Principal plan critic, adversarial evaluator, contract integrity validator, and architectural quality gate.
How to install Critic Agent
Installs to ~/.claude/agents/kaivian-it-agent-team-critic-agent.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/Kaivian/it-agent-team/HEAD/agents/critic-agent.md -o ~/.claude/agents/kaivian-it-agent-team-critic-agent.md Restart Claude Code, or start a new session, for it to be picked up.
What Critic Agent does
name: critic-agent description: Principal plan critic, adversarial evaluator, contract integrity validator, and architectural quality gate.
Critic Agent
1. Overview and Mission
The Critic Agent serves as an adversarial evaluator and architectural quality gate. Operating under the assumption that plans contain hidden defects, the Critic rigorously scrutinizes technical specifications against a 6-pillar evaluation matrix before any code generation is permitted to commence.
2. Ro
Alternatives in Development
- Dast Devils Advocate — Adversarial validator for DAST findings 812 ★
- Adversarial Plan Validator 127 ★
- Evaluator Evm Agent — EVM smart-contract vulnerability evaluator — spawned per smart_contract surface, scaffolds and runs Foundry te 97 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Sap Critic
SAP quality gate — functional specification review, configuration validation, and implementation plan critique
Principal Developer
Development Department Lead. Creates technology-aware implementation plans, assigns work to specialists, and s
Principal Agent
Lead Agent of the execution pipeline. Use this agent to draft and negotiate the Implementation Plan with the u
Section Evaluator
Evaluates completed section implementations against their section plan's Verification Criteria and Evaluator C
Product Critic
Adversarial business validation of a product or feature idea BEFORE any build effort is committed. Use PROACTI
Parallel Hunt Fix Gate
Adversarial fix gate for a /parallel-hunt batch — tries to refute each fix and either verifies it or sends it
Related Skills
Adversarial Dev
GAN-inspired three-agent harness that pits a generator against an adversarial evaluator to build applications
Bandit
The agent factory — AI coding agents on a kanban board, an adversarial critic enforcing quality, verification-
Da
Explicit Diablo invocation — adversarial critic for spec, plan, or implementation. Outputs FATAL/SERIOUS/SUSPI
