admin/openhuman
1
0
Fork 0
mirror of https://github.com/tinyhumansai/openhuman.git synced 2026-07-27 21:08:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

Make subagents async and reusable by default (#3887)

Browse Source
This commit is contained in:
Steven Enamakel
2026-06-22 15:51:48 -07:00
committed by GitHub
parent ccee419d86
commit 2d71eb3ab1
30 changed files with 4190 additions and 77 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/DELEGATION_POLICY.md
+6 -6
View File
@@ -14,18 +14,18 @@ Apply when: the task needs a tool but not specialised execution (time lookup, me
Cost: 1 tool call + parse overhead (~200-400 tokens).
Rule: prefer `current_time`, `cron_*`, `memory_*`, `memory_tree`, `read_workspace_state`, `composio_list_connections`, `ask_user_clarification`.
## Tier 3 — Spawn a sub-agent (inline)
## Tier 3 — Delegate to a reusable async sub-agent
Apply when: the task requires specialised execution (writing code, crawling docs, running shell, calling an external integration) that the orchestrator cannot do directly.
Cost: full sub-agent turn (~1-5k tokens depending on archetype).
Rule: spawn the narrowest archetype that can complete the task. Prefer inline spawn (`spawn_worker_thread` with no dedicated thread) for tasks that complete in <5 turns.
Rule: spawn the narrowest archetype that can complete the task. `spawn_subagent` is reusable and asynchronous by default: it first looks for a compatible durable worker for the same parent thread, agent id, toolkit/model/sandbox/action-root shape, and deterministic task key. If one is running, the new instruction is steered into it; if one is idle, it resumes from saved history; otherwise a new durable worker session is created.
## Tier 4 — Spawn a dedicated worker thread
Apply when: the task is long (>5 turns estimated), produces a large transcript, or the user explicitly wants it tracked as a separate thread.
## Tier 4 — Explicit worker lifecycle control
Apply when: the task is long (>5 turns estimated), produces a large transcript, needs manual inspection, or the user explicitly wants it tracked/closed separately.
Cost: same as Tier 3 but the parent thread is not flooded.
Rule: use `spawn_worker_thread` and surface a brief summary back to the parent. Do not chain workers (workers cannot spawn workers).
Rule: use `list_subagents`, `steer_subagent`, `wait_subagent`, and `close_subagent` with `subagent_session_id` for durable control. Use `fresh: true` only when the prior worker is materially incompatible or the user asks for a clean worker. Use `blocking: true` only when the parent must synchronously wait for the child result before replying. Do not chain workers (workers cannot spawn workers).
## Anti-patterns to avoid
- Spawning a sub-agent to answer a question the orchestrator already has context for.
- Delegating a tool call to a sub-agent when `current_tier <= 2` applies.
- Using `spawn_subagent` when `delegate_{archetype}` covers the task — `delegate_*` tools carry the full archetype definition and have correct tool filtering pre-configured.
- Repeatedly forcing fresh sub-agents for the same logical job, which loses worker-local context and cache affinity.
- Passing the entire parent conversation as context to a sub-agent — pass only the task-relevant slice.