Skip to content
Skill

octoclaw

by guanbear

AI Summary

This skill describes the current runtime line. It intentionally does not describe the old Python , completion-file finalizer, or JSON delivery-outbox runtime. Those were historical mechanisms and must not be

Install

Copy this and paste it into Claude Code, Cursor, or any AI assistant:

I want to install the "octoclaw" skill in my project.

Please run this command in my terminal:
# Install skill into your project
mkdir -p .claude/skills/OctoClaw && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/OctoClaw/SKILL.md "https://raw.githubusercontent.com/guanbear/OctoClaw/v0.6.0/SKILL.md"

Then restart Claude Code (or reload the window in Cursor) so the skill is picked up.

Description

OctoClaw is the TypeScript-first policy, delegation, status, and operator layer for OpenClaw. Use it when a turn should be routed through reply/delegate policy, WorkContract-backed delegation, runtime status, IM delivery, or feedback-loop tooling.

OctoClaw Runtime Skill

This skill describes the current v0.6.0 runtime line. It intentionally does not describe the old Python octoclaw_spawn.py, completion-file finalizer, or JSON delivery-outbox runtime. Those were historical mechanisms and must not be reintroduced as compatibility paths.

Current Authority

| Layer | Authority | |------|-----------| | live route | reply | delegate only | | execution lifecycle | OpenClaw native TaskFlow / sessions_spawn / native announce | | OctoClaw metadata | SQLite runtime ledger: WorkContract, route seal, native refs, spawn intent, audit events | | status display | generated projection from metadata + native lifecycle | | task-state.json | rebuildable read-model cache, not truth | | replay log | observability and evaluation evidence, not runtime truth |

Main-Agent Rules

• For direct answers, use reply; do not create work. • For delegated work, call octoclaw_dispatch; do not hand-write spawn args. • After native spawn is accepted, confirm with octoclaw_dispatch_confirm. • For status/provenance/follow-up questions, use octoclaw_status or octoclaw_task_action; do not dispatch new work just to explain old work. • Do not call or recreate octoclaw_spawn. • Do not use completion files, child-finalizer recovery, or JSON delivery outbox as the normal final-delivery mechanism. • Do not treat task-state.json as durable truth. Missing or corrupt task-state must be rebuilt or reported degraded. • Do not inject raw child transcripts into the parent context.

Tool Use

| Tool | Use | |------|-----| | octoclaw_route_hint | Ask the runtime for route guidance before model/tool escalation. | | octoclaw_policy_decide / octoclaw_route | Resolve policy and route facts. | | octoclaw_dispatch | Create a WorkContract-backed delegation plan and native spawn intent. | | octoclaw_dispatch_confirm | Bind accepted native runId / childSessionKey evidence back to OctoClaw metadata. | | octoclaw_task_action | Read task details, queue, artifacts, or supported task operations. | | octoclaw_status | Render status/details/queue/timeline from projection. |

Discussion

0/2000
Loading comments...

Health Signals

MaintenanceCommitted 2d ago
Active
AdoptionUnder 100 stars
61 ★ · Niche
DocsREADME + description
Well-documented

GitHub Signals

Stars61
Issues0
Updated2d ago
View on GitHub
MIT License

My Fox Den

Community Rating

Sign in to rate this booster

Works With

Claude Code