Agent Skills by strategicmilk
Showing 41 vetted skills indexed across 1 GitHub repositories.
Comprehensive Code Review
Perform a five-pass code review and return a structured JSON report.
Performance Review
Analyze Python functions for performance anti-patterns and generate a JSON report.
Test Coverage Analysis
Identify untested code paths and generate test specifications with a JSON report.
Best Practices Enforcement
Enforce Vetinari conventions and output remediation guidance during code reviews.
Security Audit
Review code and configuration for OWASP Top 10 vulnerabilities with CWE mapping.
Code Simplification
Detect dead code, over-abstraction, and indirection in codebases.
Scope Analysis
Generate a JSON scope report of direct and transitive code change impacts.
Requirement Specification
Generate Given-When-Then acceptance criteria from ambiguous user requests.
Effort Estimation
Predict per-task effort, token usage, and cost within Foreman task plans.
Task Decomposition
Convert ambiguous goals into directed acyclic graphs of atomic executable tasks.
Replan on Failure
Generate updated DAG plans after task failures or budget overruns.
Git Archaeology
Trace git history to uncover design decisions and regressions.
Codebase Exploration
Map codebases via LSP symbol navigation and return structured JSON reports.
Continuous Improvement
Implements PDCA cycles to analyze Vetinari episode_memory and prioritize operational improvements.
Infrastructure Research
Analyze infrastructure configurations and produce a structured report with prioritized recommendations.
Error Handling Hardening
Replace bare except clauses with specific exception types in Python codebases.
Risk Assessment
Generate risk assessment reports with probability, impact, and mitigation strategies in JSON.
Error Recovery
Classify worker task errors and apply recovery actions like retries or reassignment.
Bug Diagnosis and Fix
Diagnose and fix software bugs using a Reproduce-Localize-Root-cause-Fix-Verify workflow.
Documentation Generation
Generate API references, architecture diagrams, and changelogs from source code.
Intelligent Refactoring
Perform AST-aware refactoring of Python symbols with dry-run validation and test verification.
Code Generation
Generate Python boilerplate for agent modes, Flask endpoints, pytest tests, and dataclasses.
Lateral Exploration
Generate ranked cross-domain analogical proposals with validation plans for technical problems.
Feature Implementation
Implement fully specified Vetinari features with type safety, tests, and documentation.