critic — evidence-first critic, codereview-pilot, community, evidence-first, ide skills, incident-diagnosis, root-cause-analysis, Claude Code, Cursor, Windsurf

v1.0.0
GitHub

About this Skill

Perfect for Diagnostic Agents needing advanced incident analysis and quality gating capabilities. Evidence-first incident diagnosis. Root cause in 16–30s. Project 1.0.0 complete (29/29 checkpoints).

# Core Topics

jimmymalhan jimmymalhan
[1]
[0]
Updated: 3/20/2026

Agent Capability Analysis

The critic skill by jimmymalhan is an open-source community AI agent skill for Claude Code and other IDE workflows, helping agents execute tasks with better context, repeatability, and domain-specific guidance. Optimized for evidence-first, incident-diagnosis, root-cause-analysis.

Ideal Agent Persona

Perfect for Diagnostic Agents needing advanced incident analysis and quality gating capabilities.

Core Value

Empowers agents to automate quality gating and output contract validation end-to-end, utilizing confidence thresholds and field presence checks, with support for JSON output and gate definition using models like haiku and sonnet.

Capabilities Granted for critic

Automating incident diagnosis with root cause analysis
Validating output contracts for completeness and confidence
Generating approval or rejection decisions based on gate checks

! Prerequisites & Limits

  • Requires confidence score >= 0.70 for approval
  • Demands all 6 required fields be present and non-empty
  • No partial approvals - only APPROVED or REJECTED decisions
Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

critic

Evidence-first incident diagnosis. Root cause in 16–30s. Project 1.0.0 complete (29/29 checkpoints).

SKILL.md
Readonly

Phase 1: DISCOVER

Sub-Agent: OutputCollector (model: haiku)

  • Tools: Read
  • Prompt: Collect final output from pipeline. Check which of the 6 required fields are present: root_cause, evidence, fix_plan, rollback, tests, confidence.
  • Output: { fields_present[], fields_missing[], confidence_value }
  • Gate: output collected

Phase 2: PLAN

Sub-Agent: GateDesigner (model: sonnet)

  • Prompt: Define pass/fail for each field. Field is valid if: non-empty, has concrete content (not placeholder), confidence >= 0.70.
  • Output: { gates[{field, present, valid, reason}] }
  • Gate: 6 gates defined

Phase 3: IMPLEMENT

Sub-Agent: GateChecker (model: haiku)

  • Prompt: Check each gate. Is field present? Is it non-empty? Is confidence >= 0.70? Are all 6 fields present?
  • Output: { gates_checked[], all_pass: boolean, blocking[] }
  • Gate: all 6 gates checked

Phase 4: VERIFY

Sub-Agent: DecisionMaker (model: haiku)

  • Prompt: APPROVED if all 6 gates pass AND confidence >= 0.70. REJECTED with specific blocking issues listed. No partial approvals ever.
  • Output: { decision: "APPROVED"|"REJECTED", blocking[], confidence }
  • Gate: decision made

Phase 5: DELIVER

Sub-Agent: CriticReporter (model: haiku)

  • Prompt: Output final verdict. If REJECTED, list exactly which fields are missing/invalid and what would fix them.
  • Output: { approved: boolean, blocking[], fix_suggestions[], confidence }
  • Gate: verdict delivered

Contingency

IF output has 0 of 6 fields → REJECTED immediately. Do not attempt to fill in missing fields. IF confidence is exactly on boundary (0.70) → APPROVED but flag as "marginal — add more evidence."


Critic Role Skill

Purpose: Automate quality gating and output contract validation end-to-end.

Role: Quality Gate in the 4-agent diagnosis pipeline.

Create → Handle → Run (E2E)

Create

  • Define quality gates (confidence >= 0.70, all fields present)
  • Add blocking rules for missing/partial output
  • No partial approvals—APPROVED or REJECTED only

Handle

  • Receive: full output contract (root cause, evidence, fix plan, rollback, tests, confidence)
  • Check: confidence >= 0.70
  • Check: all 6 fields present and non-empty
  • Output: APPROVED or REJECTED with specific blocking issues

Run

  • Full pipeline: node src/local-pipeline.js or UI submit
  • Verify Critic blocks when confidence < 0.70
  • Verify Critic blocks when any field missing
  • Verify APPROVED only when all gates pass

Output Contract (Required Fields)

  1. Root Cause – Why the incident occurred
  2. Evidence – File:line citations from repo
  3. Fix Plan – Specific steps to resolve
  4. Rollback Plan – How to undo the fix
  5. Tests – How to verify the fix
  6. Confidence – Score >= 0.70

Constraints

  • ✓ Block if confidence < 0.70
  • ✓ Block if any field missing
  • ✓ List specific blocking issues when REJECTED
  • ✗ No partial approvals
  • evidence-proof – Confidence backed by tests
  • backend-reliability – Error handling for gate logic

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is critic?

Perfect for Diagnostic Agents needing advanced incident analysis and quality gating capabilities. Evidence-first incident diagnosis. Root cause in 16–30s. Project 1.0.0 complete (29/29 checkpoints).

How do I install critic?

Run the command: npx killer-skills add jimmymalhan/codereview-pilot/critic. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for critic?

Key use cases include: Automating incident diagnosis with root cause analysis, Validating output contracts for completeness and confidence, Generating approval or rejection decisions based on gate checks.

Which IDEs are compatible with critic?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for critic?

Requires confidence score >= 0.70 for approval. Demands all 6 required fields be present and non-empty. No partial approvals - only APPROVED or REJECTED decisions.

How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add jimmymalhan/codereview-pilot/critic. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use critic immediately in the current project.

Related Skills

Looking for an alternative to critic or another community skill for your workflow? Explore these related open-source skills.

View All

widget-generator

Logo of f
f

f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.

149.6k
0
AI

flags

Logo of vercel
vercel

flags is a Next.js feature management skill that enables developers to efficiently add or modify framework feature flags, streamlining React application development.

138.4k
0
Browser

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
AI