CriticSkill

Evaluates outputs for errors, hallucinations, and compliance with system invariants.

Updated Feb 25, 2026

Search Tags:#security #verification #compliance #quality gate #output validation #critic

Ready to run

npx skills add https://github.com/PixnBits/SeedClaw --skill criticskill

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: CriticSkill
Source: https://github.com/PixnBits/SeedClaw/tree/main/src/skills/agent/critic
Command: npx skills add https://github.com/PixnBits/SeedClaw --skill criticskill

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill rigorously evaluates outputs from other skills, detecting errors, hallucinations, and compliance violations to maintain high quality and adherence to system invariants.

Core Features & Use Cases

Output Verification: Critically assesses code, plans, and responses for accuracy, safety, and usefulness.
Error Detection: Identifies logical inconsistencies, security vulnerabilities, and prompt injection residues.
Structured Feedback: Provides detailed issue lists, severity scores, and concrete suggestions for improvement.
Use Case: After the coder skill generates a new script, the critic skill reviews it for security flaws, adherence to coding standards, and correct implementation of requirements before it's registered.

Quick Start

Critique the provided code snippet for accuracy and security vulnerabilities.

Frequently Asked Questions about CriticSkill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I verify AI outputs for hallucinations and compliance violations?▼

Detect security vulnerabilities in generated code by applying heuristic evaluation and rule-based critique. This identifies logical inconsistencies, prompt injection residues, and architectural compliance issues before the code is registered or deployed.

How do I enforce least-privilege principles and architectural compliance on automated outputs?▼

Enforce least-privilege principles and architectural compliance by scoring outputs across multiple axes using structured critique. The evaluation ensures generated plans and code strictly adhere to network policies and system invariants.

What is the best way to set up a quality gate for code generation pipelines?▼

Set up a quality gate for code generation pipelines by applying structured critique and severity scoring to generated scripts. This catches errors and compliance violations before the code is registered, maintaining high system quality.

Can I persist critique patterns for future output validation?▼

Persist critique patterns for future output validation by optionally saving the structured feedback and scoring results. This allows consistent application of error detection and compliance rules across subsequent skill outputs.

How do I identify prompt injection residues in automated responses?▼

Identify prompt injection residues in automated responses by critically assessing the output for security vulnerabilities and compliance violations. Heuristic methods detect these residues to maintain output safety and usefulness.