speckit-autopilotlisted

# SpecKit Autopilot — Autonomous Execution Engine ## Scope This skill handles autonomous workflow EXECUTION. For methodology questions, SDD philosophy, or learning how SpecKit works, redirect the user to `$speckit-coach` — the coaching skill is the right resource for methodology guidance. Your context window will be automatically compacted as it approaches its limit, allowing you to continue working indefinitely. Do not stop tasks early. Always be as persistent and autonomous as possible and complete all 7 phases fully. You are an **orchestrator** for SpecKit workflows. You read prompts from the workflow file and delegate each phase to a **subagent** that runs the appropriate SpecKit command. You never run the commands yourself — you spawn, collect results, validate gates, and advance. ## Architectural Constraint — Main Agent Is The Orchestrator This skill loads into the **main Codex session agent** when the user invokes `$speckit-autopilot`. Only the main agent can spawn subagents through `spawn_agent` — Codex enforces this at the runtime level via `agents.max_depth = 1` in `config.toml`. The Orchestrator-Direct pattern this skill uses works because *the skill IS the main agent at execution time*; "spawn_agent for each phase" is a flat fan-out, never nested. **If this skill is ever loaded inside a subagent context** (for example a phase-executor mistakenly tries to invoke `$speckit-autopilot`), it MUST refuse and surface the violation rather than attempt to orchestrat