Security & Testing Agent Skills

Root Cause Tracing & Debugging

Systematically traces execution errors back through the call stack to identify and fix original triggers rather than surface symptoms.

Python Testing Patterns

Implements comprehensive Python testing strategies using pytest, mocking, and test-driven development best practices.

AI Data Poisoning Red Teaming

Tests AI training pipelines for data poisoning vulnerabilities and backdoor injection to ensure model integrity.

Response Validator

Validates skill responses against the standard FABER JSON schema to ensure workflow integrity.

API Testing & Quality Assurance

Implements comprehensive API testing strategies including unit, integration, contract, and load testing patterns to ensure robust backend services.

Verification Before Completion

Enforces rigorous evidence-based verification before allowing success claims, commits, or pull requests.

Solidity Security

Hardens Ethereum smart contracts against vulnerabilities using industry-standard security patterns and gas optimization techniques.

Systematic Debugging

Eliminates trial-and-error bug fixing by enforcing a rigorous four-phase root cause analysis and verification process.

Test-Driven Development (TDD)

Enforces a strict Red-Green-Refactor workflow to ensure high-quality, verified code through test-first implementation.

Functionality Audit

Validates code correctness through isolated sandbox execution and systematic verification of outputs against expected behavior.

Blueprint Code Validator

Audits your codebase for consistency against documented specifications, architectural decisions, and design patterns.

STDD Rules

Enforces spec- and test-driven development workflows by ensuring project initialization and adherence to defined specifications.

TypeScript Testing Patterns

Generates robust unit and integration tests for TypeScript and JavaScript using modern frameworks like Jest and Vitest.

Sandbox Configuration

Manages Claude's sandboxing and isolation features for secure code execution and environment management.

Safe Solidity Refactor

Refactors non-upgradeable Solidity contracts to improve clarity and maintainability while preserving identical execution behavior and interfaces.

Protocol Audit Prep

Generates comprehensive, auditor-facing documentation and evidence bundles for DeFi protocol vaults to streamline the security review process.

Security Reviewer

Conducts comprehensive security audits of codebases, focusing on OWASP Top 10, secrets management, and AI/MCP security patterns.

Scaffold Solidity Contract

Generates audit-ready Solidity contract skeletons with comprehensive NatSpec documentation and standardized structures for DeFi development.

Skill Eval Optimizer

Optimizes and triages failed skill-loading evaluations by applying targeted fixes and executing strict retest cycles.

Skill Activation Testing

Validates and refines Claude Code skill descriptions through systematic testing to ensure accurate activation and minimize discovery errors.

TDD & Tidy First Workflow

Guides development using Kent Beck’s Test-Driven Development and Tidy First principles to ensure clean, robust, and maintainable code.

Testing Strategy Framework

Implements comprehensive testing strategies across the development lifecycle using the test pyramid and industry-standard patterns.

Time Travel Debugging with Undo

Debugs Linux programs by recording execution and traveling backwards through history to resolve intermittent failures and race conditions.

Gemini Sandbox Configuration

Configures secure sandboxing and process isolation for Gemini CLI using Docker, Podman, and macOS Seatbelt.

Quality Checklist Generator

Generates dynamic, domain-specific quality validation checklists for technical specifications to ensure requirement completeness.

Security Environment Manager

Enforces secure handling of environment variables and secrets by preventing credential leaks in code repositories.

Permission Management

Guides the configuration and troubleshooting of Claude Code's security permissions, modes, and tool-specific access rules.

Systematic Debugging

Enforces a rigorous, root-cause-first debugging methodology to eliminate guesswork and prevent regressions during the troubleshooting process.

OpenSpec Implementation Verifier

Validates code implementations against change specifications, tasks, and design documents to ensure completeness and correctness.

AI Security & Safety Benchmarks

Evaluates AI model security, robustness, and safety using standardized datasets like HarmBench, JailbreakBench, and AdvGLUE.