spae-spawn

Orchestrates build, check, and fix subagents in a loop driven by STATE.json phase transitions.

1|Updated May 12, 2026
One-click install
npx skills add https://github.com/mystilleef/spae-framework --skill spae-spawn-mystilleef
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: spae-spawn
Source: https://github.com/mystilleef/spae-framework/tree/main/skills/spae-spawn
Command: npx skills add https://github.com/mystilleef/spae-framework --skill spae-spawn-mystilleef

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Manually invoking each phase agent of the SPAE workflow (build, check, fix) for every atomic task is repetitive and error-prone. This Skill automates that loop by reading the shared STATE.json file and dispatching the correct subagent until every task is done. ## Core Features & Use Cases - Phase-driven dispatch: Reads .spae/current/STATE.json and spawns the subagent matching the current phase (build, check, or fix) with no arguments. - Stall and failure detection: Halts immediately if the phase does not change between iterations, if a subagent fails, crashes, or times out, or if STATE.json is malformed. - Strict orchestration boundaries: Never writes project files, never commits, never runs subagents in parallel, and only invokes the permitted build, check, and fix agents. - Use Case: After running the spec, plan, and inspect phases of a SPAE workstream, invoke this Skill to autonomously execute all remaining atomic tasks, cycling build, check, and fix until the workstream reaches the verify phase. ## Quick Start Run the SPAE spawn orchestrator to autonomously execute all remaining tasks in the current workstream until the state reaches the verify phase.

Frequently Asked Questions about spae-spawn

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate the SPAE build phase across multiple tasks?▼

Invoke the spae-spawn skill after the inspect phase completes. It reads STATE.json, spawns the subagent matching the current phase, and repeats the loop until every task is marked done and the phase reaches verify.

How does the spawn orchestrator decide which subagent to run?▼

It uses a fixed phase-agent map: the build phase spawns the build agent, check spawns check, and fix spawns fix. The phase value in STATE.json is re-read before every dispatch, and the loop exits when the phase becomes verify.

What happens if a subagent fails or the phase stops changing?▼

Execution halts immediately. The orchestrator stops on any Failed status, crash, timeout, or AgentError, and also halts if STATE.json shows the same phase at the end of an iteration as at its start, preventing infinite loops.

Can the spawn orchestrator modify project files or STATE.json?▼

No. The orchestrator is strictly read-only with respect to project files and STATE.json. All state transitions are performed by the build, check, and fix subagents it invokes; it only reads state and reports results.

When should I not use the SPAE spawn orchestrator?▼

Do not use it before the plan and inspect phases have populated STATE.json with tasks and an active cursor, or when you need manual review between each phase. It is designed for autonomous execution of an already-planned workstream.