deeprefine
Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.
npx skills add HKUST-KnowComp/DeepRefine-Skill --skill deeprefine_skill --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# DeepRefine — Agent refinement loop (strict) ## Default safety policy: dry-run only A normal `/deeprefine` invocation **MUST NEVER** call `deeprefine apply`. The default `/deeprefine` workflow must stop after: 1. `deeprefine loop validate` 2. `deeprefine review` 3. showing the proposed actions and HIGH/MEDIUM/LOW review report to the user Then ask the user for explicit approval. Only if the user's **next message** explicitly says to approve/apply/write the graph may you run: ```bash deeprefine apply --refresh-wiki --trace-file ... --refinement-file ... deeprefine loop finish --trace-file ... --refinement-file ... ``` Do not treat generation of `<refinement>` actions as approval. Do not treat a valid trace as approval. Do not apply in the same `/deeprefine` turn. --- You **MUST** implement the **same control flow** as `DeepRefine.refine()` in DeepRefine (`autorefiner/src/deeprefine.py`). | Component | Agent mode | CLI `deeprefine refine` | |-----------|------------|-------------------------| | Retrieval | `graphify query` + k-hop from `graph.json` | FAISS retriever | | LLM | **Your session model** | External API / vLLM | | Graph writes | Dry-run proposal + `deeprefine review`; `dee
- Default safety policy: dry-run only
- FORBIDDEN (hard stop)
- Constants (match refinerunner.py / DeepRefine)
- Mandatory artifact
- Query queue selection (default behavior of /deeprefine)
- Control flow (must match DeepRefine.refine())
- Evidence-aware review rules
- LLM prompts (verbatim — do not paraphrase)
- Judgement (answerablejudgement)
- Error abduction (errorabduction) — only if len(interactionhistory) > 1
- KG refinement actions (kgrefinementaction) — only if len(interactionhistory) > 1
- Per-query checklist (report in chat)
- Commands (in order)
- looptrace.json schema
deeprefine apply --refresh-wiki --trace-file ... --refinement-file ... deeprefine loop finish --trace-file ... --refinement-file ... deeprefine loop init --query "<exact question>" deeprefine review --trace-file graphify-out/.deeprefine/loop_trace_<id>.json --refinement-file graphify-out/.deeprefine/refinement_actions_<id>.txt mkdir -p graphify-out/.deeprefine cp graphify-out/graph.json graphify-out/.deeprefine/graph.json.bak deeprefine history sync-memory deeprefine history list --pending deeprefine loop init --query "<question>" deeprefine loop validate --trace-file graphify-out/.deeprefine/loop_trace_<id>.json
What does the deeprefine skill do?
Agent-native DeepRefine refinement loop — same control flow as DeepRefine.refine(), graphify search instead of FAISS, session LLM, dry-run review before approved graph writes.
How do I install it?
Run `npx skills add HKUST-KnowComp/DeepRefine-Skill --skill deeprefine_skill --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From HKUST-KnowComp/DeepRefine-Skill, a repository with 89 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.