LLMSkillHub
mirroredCopied from 7df-lab/devo under MIT. 7df-lab has not claimed this page, and is not affiliated with LLM SkillHub.

7df-lab / refine

Continual harness refinement (prompt notes, memory, skills, subagents). Schedule `await refine.run(...)` / `await refine.status()`; apply happens at turn boundary.

0
SKILL.md
785 B

Refine skill

Schedule continual harness refinement from the Python REPL. Refinement never runs mid-cell — the host applies pending edits when the turn ends.

await refine.status()
await refine.run(instructions="Prefer local memory for this session's blockers")
await refine.run(instructions="Promote durable preference", global_=True)

Harness CRUD for memories and related entries lives on rlm.harness (requires
RLM_SESSION_DIR / RLM_HARNESS_STATE_DIR from the host):

rlm.harness.create_memory(title="...", content="...")
rlm.harness.overview()