Skip to content

Reference

All skills conform to the Gold Standard SKILL.md v1.0 specification. Every skill includes: Blocking Violations table, Verification with quality gates, Performance & Cost model, structured Examples, Anti-Patterns tables, and full References/Changelog.

Skills

Core Workflow (9)

Skill Description
brainstorming Structured exploration before committing
writing-plans Detailed step-by-step implementation plans
executing-plans Disciplined execution with dependency graph resolution
test-driven-development Red-Green-Refactor cycle
debugging Rapid Fix + Systematic root cause analysis
requesting-code-review Structured review requests with severity tags
receiving-code-review Handle feedback systematically
finishing-a-development-branch Clean merge, branch cleanup, release notes
using-git-worktrees Parallel development on multiple branches

Domain Expert (30)

Architecture

Skill Description
tech-lead Architecture, planning, tech debt assessment
frontend-architect Component design, state management, performance
backend-architect API design, service boundaries, data flow
infra-architect Infrastructure design, scalability, DR

Development

Skill Description
code-polisher Refactoring, dead code removal, consistency
migration-upgrader Dependency upgrades, breaking change management
mobile-architect Platform-specific design, offline, performance
legacy-archaeologist Understand and safely modify legacy code
python-patterns Typing, async, packaging, testing idioms

Testing

Skill Description
test-genius Unit test design, edge cases, testability
e2e-test-specialist Playwright/Cypress, user journeys, flakiness
eval-harness AI feature evaluation with Pass@K metrics

DevOps

Skill Description
ci-config-helper Pipeline configuration, caching, matrix builds
docker-expert Multi-stage builds, security, compose patterns
k8s-orchestrator Deployments, services, ingress, autoscaling
observability-specialist Metrics, traces, logs, alerting, dashboards

Data & AI

Skill Description
data-engineer Schema design, ETL patterns, data quality
data-analyst Statistical validation, visualization, A/B testing
ml-engineer MLOps pipeline, model evaluation, bias detection
search-vector-architect Embedding models, hybrid search, RAG
database-migrations Expand-Migrate-Contract, zero-downtime, rollback
regex-vs-llm-structured-text Choosing between regex and LLM for extraction

Security & Performance

Skill Description
security-reviewer OWASP Top 10, injection, secret detection
performance-profiler CPU, memory, network profiling and optimization

Documentation & UX

Skill Description
doc-writer API docs, READMEs, architecture documentation
ux-designer Accessibility, responsive design, design systems

Product & Meta

Skill Description
product-manager Feature planning, stakeholder communication
workflow-orchestrator Step functions, retry, compensation logic
skill-generator Create new SKILL.md files with proper structure
multi-plan Multi-model collaborative planning synthesis
multi-execute Parallel prototyping + refactoring + audit

Meta & Learning (12)

Skill Description
continuous-learning-v2 Instinct extraction and evolution from sessions
search-first Research-before-coding discipline
autonomous-loops Multi-step AI pipeline patterns
skill-stocktake Quality audit and library maintenance
cost-aware-llm-pipeline Model routing and token budget tracking
verification-loop Tiered verification (Surface / Standard / Deep)
iterative-retrieval Progressive context refinement for subagents
content-hash-cache-pattern SHA-256 caching for file processing
plankton-code-quality Write-time formatting and linting enforcement
autoresearch-loop Karpathy-style self-improvement eval loop

Token Optimization (2)

Skill Description
rtk CLI proxy for 60-90% token savings
context-optimizer Context offloading and session continuity

Agent Personas

Category Persona Purpose
Development tech-lead Architecture, planning, tech debt assessment
planner Z-Pattern decomposition, dependency mapping
architect System design, ADRs, trade-off analysis
Quality & Testing code-reviewer Code quality, severity-tagged feedback
qa-engineer Test strategy, coverage gates, flakiness prevention
accessibility-auditor WCAG A/AA/AAA, screen reader, contrast, ARIA
adversarial-tester Chaos experiments, fuzz testing, race conditions
Security & Compliance security-reviewer OWASP Top 10, secret detection, CVE audit
data-privacy-officer GDPR/CCPA/SOC2, PII discovery, consent, retention
Operations & Releases incident-commander SEV1-4 classification, mitigation, postmortem
release-manager Semver, changelog, rollout strategy, monitoring

Workflow Chains

Feature Development

search-first → tech-lead → brainstorming → writing-plans →
test-driven-development → executing-plans → verification-loop →
requesting-code-review → finishing-a-development-branch →
continuous-learning-v2

Bug Fix

debugging → test-driven-development →
verification-loop → finishing-a-development-branch →
continuous-learning-v2

Incident Response

incident-commander → [Mitigation] → debugging →
test-driven-development → verification-loop →
finishing-a-development-branch

Release

release-manager → verification-loop →
finishing-a-development-branch → observability-specialist →
continuous-learning-v2

High-Complexity

search-first → architect → multi-plan → [Approval] →
multi-execute → verification-loop → security-reviewer →
finishing-a-development-branch

Accessibility Audit

accessibility-auditor → [Fixes] → verification-loop →
requesting-code-review → finishing-a-development-branch

Adversarial Test

adversarial-tester → [Chaos/Fuzz] → debugging →
executing-plans → verification-loop →
finishing-a-development-branch

Commands

Command Skill Purpose
/mega-mind mega-mind Orchestrator entry point
/brainstorm brainstorming Explore approaches
/plan writing-plans Create implementation plan
/execute executing-plans Execute plan with tracking
/debug debugging Debug systematically
/review requesting-code-review Request code review
/ship finishing-a-development-branch Deploy to production
/tdd test-driven-development Test-first development
/verify verification-loop Verify before marking done
/mega-mind route <request> mega-mind Analyze and route
/mega-mind status mega-mind Show session state