session-reflection

jwilger

Structured session analysis and system prompt refinement. Analyzes user interventions, categorizes process gaps, generates project-specific system prompts, and maintains working state for crash recovery. Activate after milestones, repeated corrections, session restarts, or on user request.

Deep Research

jwilger

Conduct enterprise-grade research with multi-source synthesis, citation tracking, and verification. Use when the user needs comprehensive analysis requiring 10+ sources, verified claims, or comparison of approaches. Triggers include "deep research", "comprehensive analysis", "research report", "compare X vs Y", or "analyze trends". Do NOT use for simple lookups, debugging, or questions answerable with 1-2 searches.

refactoring

jwilger

Safe refactoring patterns under green tests. Extract, rename, move, simplify -- always in a separate commit from behavior changes. Activate when cleaning up code after TDD GREEN, during code review rework, or when the codebase needs structural improvement. Triggers on: "refactor", "clean up", "extract method", "rename", "simplify".

pipeline

jwilger

Autonomous build-phase orchestrator. Manages slice queue, TDD pair dispatch, full-team code review, mutation testing, CI integration, and auto-merge with quality gates. Replaces manual coordinator overhead during build phase. Activate when running factory mode with ensemble-team.

ticket-triage

jwilger

Evaluate whether a development ticket (user story, feature request, bug report, etc.) is ready for development, and provide specific, actionable feedback if it is not. Use this skill whenever the user asks to triage, evaluate, assess, review, or check the readiness of a ticket, story, issue, or work item. The ticket can come from anywhere: pasted inline, read from a file, fetched from Jira or another tracker via MCP, or any other source. Also use this when a user asks "is this ticket ready?" or "what's wrong with this ticket?" or wants to improve a ticket's specification.

agent-coordination

jwilger

Multi-agent coordination discipline: one-message-then-wait, idle heartbeat handling, no polling loops, sequential agent spawning, and proper shutdown protocol. Activate when orchestrating multiple agents, managing agent teams, or coordinating handoffs between agents.

tdd

jwilger

Adaptive test-driven development cycle. Detects harness capabilities and routes to guided (manual phase control) or automated (orchestrated) mode. Invoke with /tdd for automated or /tdd red|domain|green|commit for guided.

atomic-design

jwilger

Brad Frost's Atomic Design for UI component hierarchy: atoms, molecules, organisms, templates. Activate when building user interfaces, creating design systems, organizing components, or structuring frontend code. Applies to any UI framework (React, Vue, SwiftUI, etc.).

user-input-protocol

jwilger

Structured checkpoint format for requesting human input. When an agent needs a decision, it must stop, present context, show options, and wait. Activate when delegating to subagents, running background tasks, or hitting any decision point that requires human judgment.

debugging-protocol

jwilger

Systematic 4-phase debugging: understand the failure, form hypotheses, test one change at a time, fix with confidence. Activate when tests fail unexpectedly, errors occur, behavior is wrong, or something that worked before is now broken. Triggers on: "debug", "why is this failing", "test failure", "unexpected error", "bug", "broken".

bootstrap

jwilger

Zero-config SDLC onboarding. Detects project environment, harness capabilities, and harness type. Configures TDD mode, generates AGENTS.md, offers optional enforcement hooks, and recommends skills by phase. Single entry point for all harnesses.

factory-review

jwilger

Structures the human review experience for factory-mode builds. Audit trail summaries, PR digests, retrospective synthesis, quality trend tracking, and autonomy tuning interface. Activate during Phase 3 human review.

architecture-decisions

jwilger

Architecture Decision Records with a four-phase lifecycle (RESEARCH → DRAFT → HOLD → MERGE). Activate when making technology choices, defining system boundaries, recording architectural decisions, or creating/updating ARCHITECTURE.md. Enforces research-before-writing and explicit merge authorization.

ensemble-team

jwilger

Set up a full AI ensemble/mob programming team for any software project. Creates team member profiles (.team/), coordinator instructions (.team/coordinator-instructions.md), project owner constraints (PROJECT.md), team conventions (AGENTS.md), architectural decisions (docs/ARCHITECTURE.md), domain glossary, and supporting docs. Use when: (1) starting a new project and wanting a full expert agent team, (2) the user asks to "set up a team", "create a mob team", "set up ensemble programming", or "create agent profiles", (3) converting an existing project to the driver-reviewer mob model, (4) the user wants AI agents to work as a coordinated product team with retrospectives and consensus-based decisions.

mutation-testing

jwilger

Mutation testing to validate test quality before PR creation. Runs mutation tools, enforces 100% kill rate, reports surviving mutants with recommended fixes. Activate when validating test coverage, preparing pull requests, checking test quality, or when asked about mutation testing.

error-recovery

jwilger

Structured error classification and recovery during autonomous operation. Classify runtime errors, apply retry strategies with backoff, maintain error logs, and escalate intelligently. Activate when encountering API failures, build tool crashes, file permission issues, or unexpected runtime errors during autonomous work. Triggers on: "error recovery", "retry", "API failure", "crash recovery", "service unavailable".

event-modeling

jwilger

Event modeling facilitation for discovering and designing event-sourced systems. Four phases: domain discovery, workflow design (9-step process), GWT scenario generation, and model validation. Activate when starting a new project, designing features, modeling domains, writing Given/When/Then scenarios, or discussing event sourcing and domain-driven design.

pr-ship

jwilger

Pull request creation and merge workflow. PR title/body templates, conventional commit guidance, review checklist summary, merge strategy selection. Activate when code review passes and work is ready to ship. Triggers on: "create PR", "ship it", "merge", "open pull request", "ready to merge".

memory-protocol

jwilger

File-based knowledge persistence patterns: when to store discoveries, when to recall past solutions, and how to organize project memory. Activate when starting tasks, encountering errors, making decisions, or when context may be lost between sessions.

ci-integration

jwilger

Deterministic CI/CD interaction patterns. Push-and-wait discipline, failure triage, self-healing for lint/format/infra failures, structured output for pipeline consumption. Activate when interacting with CI/CD systems.

code-review

jwilger

Three-stage code review protocol: spec compliance, code quality, and domain integrity. Activate when reviewing code, preparing PRs, assessing implementation quality, or checking that code matches requirements. Triggers on: "review this code", "prepare PR", "check implementation", "code quality", "does this match the spec".

task-management

jwilger

Work decomposition, dependency ordering, and status tracking for software tasks. Activate when breaking down features into tasks, managing work items, tracking dependencies, creating stories or epics, or asking what to work on next. Works with any task tool: harness-native todos, dot CLI, GitHub Issues, or file-based tracking.

domain-modeling

jwilger

Domain modeling principles: parse-don't-validate, make invalid states unrepresentable, primitive obsession detection, semantic types, and domain veto authority. Activate when designing types, reviewing code for domain integrity, or when domain review is needed in a TDD cycle.

design-system

jwilger

Collaborative design system creation using Atomic Design methodology. Produces a specification artifact with philosophy, tokens, and component hierarchy. Activate when creating a design system, defining visual language, specifying UI tokens, or planning component architecture before implementation begins.