Provides a focused 4-step bugfix process (Investigate → Write failing test → Fix → Verify) and status conventions for quick, auditable fixes. It prescribes roles/agents for each step, test-type guidance (unit vs e2e), output formatting, and escalation to deeper debugging when needed.
Use when a reproducible bug is found and you want a fast, disciplined response that produces tests and verifiable fixes. Not intended for feature work or large refactors; escalate to run-orchestrator for complex cases.
Designed for developer-assistant agents (Claude Code, GitHub Copilot-like tools, or Codex-based agents) that can run tests, read diffs, and update run-state files.
A TDD-driven bugfix workflow skill that enforces root-cause investigation and failing-test-first discipline. No bundled scripts — purely instructional. Well-written process with clear steps and escalation paths, but tightly coupled to the aura-frog ecosystem (references external rules and skills not bundled). AutoInvoke with broad triggers like 'broken' or 'not working' risks unwanted activations.
Clean instructional skill, no security concerns. Main weakness is ecosystem coupling and over-broad autoInvoke triggers.
Flutter Expert
Expert Flutter/Dart patterns and best practices for building maintainable cross-platform mobile apps (state management, navigation, testing).
Documentation (ADR & Runbook)
Generates and templates operational documentation: Architecture Decision Records (ADRs) and runbooks, with templates and principles for when and how to record d
Plan Tree Validator
Validates hierarchical plan trees against 8 invariants (parent/child integrity, no orphans, valid status, monotonic revisions, DAG acyclicity, etc.) and refuses