What problem does it solve? When manual testing or implementation goes wrong, developers need to investigate logs, database state, and recent code changes without burning the primary session's context. This Skill bootstraps a focused debugging session that gathers evidence from multiple sources and produces a structured root-cause report. ## Core Features & Use Cases - Parallel Investigation: Spawns concurrent Task agents to analyze daemon and WUI logs, query the SQLite database, and inspect git state simultaneously. - Structured Debug Report: Presents findings with evidence from logs, database queries, and git diffs, plus a root-cause hypothesis and concrete next steps. - Read-Only Investigation: Examines system state without editing files, keeping the debugging session safe and isolated. - Use Case: During manual testing a session appears stuck. Invoke this Skill, describe the symptom, and it checks recent logs for errors, queries the sessions table for stuck states, and reviews uncommitted changes to pinpoint the cause. ## Quick Start Ask the agent to debug the current issue by describing what went wrong and letting it investigate the logs, database, and git state.
