Skill Claude CodeCodex
Turn the work just done into VETTED lessons that survive across sessions and projects — harvest the session's decisions/errors, draft candidate lessons, refute them adversarially, and promote only the survivors into taskmanager memories. The self-improvement loop, gated so a false lesson can never silently change how…
Skill Claude CodeCodex
The suite's risk-sized front door: given a request, name the right top-level PIPELINE (which existing entry points, in what order) by scope and risk — analysis, trivial change, bug fix, feature/high-risk, or codebase health. Advisory; it routes the pipeline only and never sizes verification depth (the brain's job).…
Skill Claude CodeCodex
Use when implementing any change in existing code — when tempted to clean up nearby code, add unrequested validation or options, fix something "arguably in scope", or when the diff is growing past what was asked.
Skill Claude CodeCodex
A durable, disposable home for in-flight thinking that overflows a single context window — layered scratch → (optional) blackboard → (optional) conclusions under thoughts/ /, refuted by the existing adversarial-verify and drained into durable truth (scribe / deep-analysis) when it settles. Trigger: ONLY when one task…
Plugin Claude Code
Interactive PRD (Product Requirements Document) builder with comprehensive interview-driven discovery, gap analysis, and taskmanager integration.
Agent
Use this agent when conducting PRD interviews, creating product requirement documents, planning new products or features, or when comprehensive discovery is needed before implementation. This agent should be triggered proactively when a user describes a product idea or feature and thorough planning would benefit…
Command
Document a bug fix with problem analysis, root cause investigation, and regression testing strategy.
Command
Create a feature-focused PRD with lighter interview process optimized for new features within existing products.
Command
Start a comprehensive PRD interview to transform an idea into a complete Product Requirements Document.
Skill Claude CodeCodex
This skill should be used when conducting PRD interviews, creating product requirements documents, planning new features, documenting bug fixes, or when using commands like "/prd-builder:prd", "/prd-builder:feature", "/prd-builder:bugfix", or "/prd-builder:refine". Provides comprehensive interview frameworks and…
Plugin Claude Code
The living docs/ single source of truth: typed, adversarially-verified documentation every plugin keeps current. Captures decisions and errors as they happen, curates them into accurate ADRs/incidents/STATUS, and verifies every claim against the code — stale docs are a defect. Bootstraps docs/ on brownfield repos via…
Agent
The scribe curator for the docs/ single source of truth. Dispatched at end of work (a Stop hook) or by scribe:sync with the FULL work context — the real git diff, the decisions and errors from the capture log, and the task that was done — to WRITE/UPDATE docs/ accurately per the documentation-discipline obligation…
Agent
Independent, adversarial, refute-first verifier of the living docs/ single source of truth against reality. Dispatched by scribe:verify and by the Stop-curate flow to check every doc claim against the code per the typed truth model — system facts vs code (drift = defect), ADR currency + consistency (an accepted ADR…
Command
Bootstrap the living docs/ single source of truth — scaffold the canonical skeleton (greenfield) or reverse-engineer it from real code via a maestro deep-analysis pass (brownfield). Idempotent; never clobbers existing docs.
Command
Consolidate recent work into docs/ then adversarially verify it — curate the capture log + git diff into ADRs/incidents/STATUS/roadmap/open-questions, then check every claim against the code.
Command
Adversarially verify docs/ against the typed truth model + cross-doc staleness — multi-vote, advisory drift report.
Skill Claude CodeCodex
Use whenever you make or discover a decision, fix a non-trivial error, change architecture or what the system does, or defer something undecided — and before acting on any task. The canonical contract for how docs/ works: docs/ is the single source of truth (read it first, write it after), its truth is typed and…
Plugin Claude Code
Plan and execute tasks from PRDs with hierarchical subtasks, dependency tracking, and project memories. Features dashboard visualization, autonomous batch execution, and persistent memory for constraints, decisions, and conventions.
Agent
Data and invariants spec for the MWGuerra Task Manager. Defines the structure and rules for .taskmanager/taskmanager.db (SQLite database) and logs. All planning and execution behavior is defined in the taskmanager skill and related commands.
Agent
Adversarial acceptance-criteria verifier for the MWGuerra Task Manager. Dispatched by taskmanager:verify and by run's completion gate to independently, refute-first confirm that a task / milestone / PRD's acceptance criteria are actually met. Assumes the work is INCOMPLETE and demands conclusive evidence. Returns a…
Command
Export SQLite database to JSON or individual task markdown files.
Command
Manage project memories (constraints, decisions, conventions) - add, list, search, update, deprecate with FTS5 full-text search.
Command
Parse a PRD into tasks, or expand existing tasks into subtasks.
Command
Research a topic using web search and codebase analysis, storing findings as memories.