Agent Skills by Patrick Mittelbach
Showing 29 vetted skills indexed across 1 GitHub repositories.
using-git-worktrees
Create isolated Git worktrees for new branches and run baseline tests.
sadd:multi-agent-patterns
Design multi-agent architectures with context isolation and coordination protocols.
test-driven-development
Guide test-first software changes through a Red-Green-Refactor loop.
systematic-debugging
Debug technical issues through four-phase root-cause investigation and failing-test-first fixes.
autonomous-loops
Orchestrate autonomous Claude Code loops for iterative software development.
customaize-agent:thought-based-reasoning
Select prompting strategies for complex multi-step reasoning tasks.
dispatching-parallel-agents
Dispatch independent debugging agents to resolve concurrent test failures.
kaizen:plan-do-check-act
Guide iterative process improvement through Plan-Do-Check-Act experiment cycles.
tech-debt
Prioritize technical debt by scoring impact, risk, and effort.
executing-plans
Execute implementation plans with review checkpoints and verification gates.
finishing-a-development-branch
Finalize a completed git branch with merge, PR, keep, or discard options.
architecture
Create or evaluate Architecture Decision Records with trade-off analysis.
architecture-designer
Guide system architecture design and ADR authoring with Mermaid diagrams.
brainstorming
Transform ambiguous product ideas into an approved design spec through staged discovery.
sadd:tree-of-thoughts
Explore, prune, and evaluate solution candidates using Tree of Thoughts reasoning.
ai-regression-testing
Build sandbox-mode API contract tests that assert response fields and parity.
writing-plans
Translate requirements into test-driven implementation plans with file paths and commit checkpoints.
requesting-code-review
Dispatch a code-review subagent over a git SHA range for severity-ranked feedback.
frontend-design
Creates distinctive frontend interfaces from a defined visual direction into functional UI code.
receiving-code-review
Verify and clarify code review feedback before incremental implementation.
writing-skills
Apply test-driven development to create and validate SKILL.md files.
reflexion:critique
Coordinate multi-perspective critique and synthesize consensus reports on completed work.
write-spec
Convert feature ideas into structured PRDs with requirements and success metrics.
mcp-builder
Guide MCP server implementation in Python or Node/TypeScript with tool schemas and XML evaluations.