Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Switched builder script from Python to Node.js: replaced build_spawn_payload.py with build_spawn_payload.mjs - Updated documentation and usage instructions to reference the new .mjs script - Removed Python references from setup and payload generation examples - Clarified that the builder is now Node.js-based in the CLI options section
What This Skill Does
Builds sessions_spawn payload JSON from reusable profile configs without executing the spawn. Profiles define model, thinking, timeout, cleanup, agentId, and label. A Node.js script merges CLI options, profile values, and defaults into ready-to-use command JSON.
Separating payload generation from execution lets you inspect and version spawn configs before any agent is actually started.
When to use it
- Preparing a subagent spawn payload for review before execution
- Reusing a named profile across multiple agent tasks
- Overriding a profile's model at the CLI for a one-off run
- Logging generated payloads to a JSONL file for auditing
- Templating subagent configs for a team workflow
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: build a spawn payload using the 'heartbeat' profile for a context summary task
- 1Reads spawn-profiles.json to load the 'heartbeat' profile
- 2Resolves model, thinking, timeout, and cleanup from profile and defaults
- 3Accepts --task and --label from CLI options
- 4Merges all values following CLI > profile > defaults priority
- 5Prints the final sessions_spawn-ready JSON to stdout
{"task":"Analyze recent context and return a compact summary","label":"heartbeat-test","model":"...","thinking":"...","runTimeoutSeconds":...,"cleanup":"..."}