AgentSkillsCN

subagent-spawn-command-builder

从 JSON 配置文件中构建 sessions_spawn 命令的载荷。当您希望拥有可复用的子代理配置文件(包括模型、思维逻辑、超时设置、清理流程、agentId 以及标签),并生成开箱即用的命令级 JSON 而无需实际执行 spawn 操作时,即可使用此功能。

SKILL.md
--- frontmatter
name: subagent-spawn-command-builder
description: Build sessions_spawn command payloads from JSON profiles. Use when you want reusable subagent profiles (model/thinking/timeout/cleanup/agentId/label) and command-ready JSON without executing spawn.

subagent-spawn-command-builder

Generate sessions_spawn payload JSON from profile config. This skill does not execute sessions_spawn; it only builds payload/command JSON.

Files

  • Profile template: state/spawn-profiles.template.json
  • Active profile config: state/spawn-profiles.json
  • Builder script: scripts/build_spawn_payload.py
  • Builder log: state/build-log.jsonl

Supported sessions_spawn parameters

  • task (required)
  • label (optional)
  • agentId (optional)
  • model (optional)
  • thinking (optional)
  • runTimeoutSeconds (optional)
  • cleanup (keep|delete, optional)

Setup

bash
cp skills/subagent-spawn-command-builder/state/spawn-profiles.template.json \
   skills/subagent-spawn-command-builder/state/spawn-profiles.json

Then edit spawn-profiles.json.

Generate payload

bash
python3 skills/subagent-spawn-command-builder/scripts/build_spawn_payload.py \
  --profile heartbeat \
  --task "Analyze recent context and return a compact summary" \
  --label heartbeat-test

The script prints JSON directly usable for sessions_spawn.

Merge/priority rule

Value resolution order is:

  1. CLI option (--model, --thinking, etc.)
  2. Profile value (profiles.<name>.*)
  3. Defaults value (defaults.*)

task always comes from CLI --task.

CLI options

Note: if generated tasks include Python execution steps, write commands with python3 (not python).

  • --profile (required)
  • --task (required)
  • --label
  • --agent-id
  • --model
  • --thinking
  • --run-timeout-seconds
  • --cleanup keep|delete