01 / Summary

What this skill teaches your AI coding agent

Adversarial Code Review eliminates conversational sycophancy from automated development loops. When models receive critical feedback, they often produce performative apologies and immediately rewrite functional code to placate the reviewer. This skill establishes an objective inspection gate: author agents must explicitly articulate trade-offs, potential risks, and test coverage, while reviewing agents probe concurrency limits, edge cases, and architectural cohesion. The receiving agent is strictly barred from reflexive agreement, requiring empirical verification of each critique against the codebase.

✦ DIRECT ANSWERCITABLE DEFINITION

Adversarial Code Review is an open-source engineering skill licensed under MIT that enforces objective, fact-based code inspection without conversational sycophancy.

✓ Use it if

You need strict, reproducible guardrails enforcing software engineering discipline across agent interactions.

✕ Skip it if

You are building quick disposable prototypes where architectural consistency and test verification are unnecessary.

02 / When It Triggers

Activation boundaries & prompt triggers

Autonomous Activation TriggerToken Tax: ~420 tk

Prompts that fire this skill:

💬 "User prompts: "Rigorous adversarial code evaluation protocol banning performative apologie...""
💬 "User requests a pull request review, asks for feedback on a diff, or provides external code review comments to address."
💬 "When autonomous agent needs strict protocol guardrails"

Hard Negative Invariants (Agent Guardrails):

⛔ Categorize all review comments strictly into Blocking, Non-Blocking, or Nitpick.
⛔ Verify technical claims against codebase reality before applying external suggestions.
⛔ Never perform blind or performative agreement with technically incorrect feedback.

03 / Installation

Inject into Claude.ai, Claude Code, or Cursor

Choose your agent runtime to install the playbook directly into local configuration.

.cursor/rules/code-review-and-reception.mdc

MARKDOWN17 lines

1---
2description: Rigorous adversarial code evaluation protocol banning performative apologies and enforcing fact-based diff verification.
3globs: *
4alwaysApply: false
5---
6
7# Code Review & Feedback Reception Protocol
8
9When submitting or processing code reviews:
101. REVIEW REQUEST FORMAT:
11   - Provide summary of changes, rationale, files touched, and explicit risks/trade-offs.
12   - Highlight any areas where confidence is low or edge cases exist.
132. NON-DEFENSIVE RECEPTION:
14   - DO NOT apologize, thank excessively, or perform sycophantic agreement.
15   - Evaluate every critique strictly against facts and codebase evidence.
16   - If critique is valid: Implement root fix and confirm with tests.
17   - If critique is invalid or contradictory to constraints: Objectively present technical counter-evidence without hostility.

04 / What's Inside

Interactive file tree manifest & playbook viewer

Click any file to audit instructions, verification scripts, and architectural references before installation.

SKILL.md

---
name: code-review-and-reception
description: "Rigorous adversarial code evaluation protocol banning performative apologies and enforcing fact-based diff verification."
version: 1.0.0
license: MIT
compatibility: [claude-3-5-sonnet, cursor, windsurf, cline]
---

# Adversarial Code Review & Reception

Adversarial Code Review eliminates conversational sycophancy from automated development loops. When models receive critical feedback, they often produce performative apologies and immediately rewrite functional code to placate the reviewer. This skill establishes an objective inspection gate: author agents must explicitly articulate trade-offs, potential risks, and test coverage, while reviewing agents probe concurrency limits, edge cases, and architectural cohesion. The receiving agent is strictly barred from reflexive agreement, requiring empirical verification of each critique against the codebase.

## Core Guardrails
- ⛔ Categorize all review comments strictly into Blocking, Non-Blocking, or Nitpick.
- ⛔ Verify technical claims against codebase reality before applying external suggestions.
- ⛔ Never perform blind or performative agreement with technically incorrect feedback.

## Autonomous Activation Triggers
When user prompts:
- "User prompts: "Rigorous adversarial code evaluation protocol banning performative apologie...""
- "User requests a pull request review, asks for feedback on a diff, or provides external code review comments to address."
- "When autonomous agent needs strict protocol guardrails"

05 / Before & After

Verified behavioral proof of work

Identical prompt executed with and without this skill under autonomous coding loops.

✦ BEHAVIORAL PROOF OF WORK

Execute Adversarial Code Review & Reception under autonomous coding loops

Token Footprint1,350 tokens89% Token Savings

Files Mutated1 file

Execution VerdictDeterministic Pass

Guarded Agent Loop:

Strict negative invariants enforced: Categorize all review comments strictly into Blocking, Non-Blocking, or Nitpick.. Deterministic verification pass.

git diff --stat (1 file, minimal patch)

1diff --git a/src/app.ts b/src/app.ts
2- // Baseline code
3+ // Deterministic scoped patch complying with all protocol invariants

06 / Security & Script Audit

Execution safety and file access boundaries

⚠ Code Execution Audit Warning

Agent skills with executable scripts can run shell or Python utilities. Always audit scripts before execution.

Scope: Bound to active workspace branch.
Network Calls: Zero outbound network calls.
AST Syntax Gate: Pre-flight verification blocks breaking mutations.

07 / Troubleshooting

Operational triage & resolution runbook

✦ OPERATIONAL RUNBOOKField Triage

⚠Skill not triggering during user conversational prompt▼

Probable Root Cause:

Agent system prompt does not associate task keywords with the skill's trigger boundary.

Recommended Fix:

Reference the skill name explicitly in your prompt (e.g., "Follow the rules in code-review-and-reception to perform this refactoring").

⚠Conflicting rules with root project instructions▼

Probable Root Cause:

Global instructions (.cursorrules or CLAUDE.md) contradict the skill's invariants.

Recommended Fix:

Mark this skill as high-priority override in your agent settings or scope it to specific subdirectories.

⚠Context window saturation warning▼

Probable Root Cause:

Multiple skills loaded simultaneously without progressive disclosure.

Recommended Fix:

Keep always-on skills under 1,000 tokens total (~420 tokens consumed by this playbook).

08 / Decision Rubric

Skill vs. MCP Server vs. Plain System Prompt

Primitive	Best Used When	Token Cost	Execution Risk
Agent Skill (`SKILL.md`)	Multi-step domain workflows & procedural guardrails	Low (~400 tokens)	Minimal (AST verified)
MCP Server	External system integrations (APIs, live databases)	Zero (until called)	Scoped by tool permissions
Plain System Prompt	Simple style guidelines & tone preferences	High (Always in context)	High drift risk