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.

91 stars 9 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add HKUST-KnowComp/DeepRefine-Skill --skill deeprefine

Manual install

git clone --depth 1 https://github.com/HKUST-KnowComp/DeepRefine-Skill.git /tmp/DeepRefine-Skill
mkdir -p ~/.claude/skills
cp -r /tmp/DeepRefine-Skill/. ~/.claude/skills/deeprefine

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 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`; `deeprefine apply` only after user approval | Dry-run by default; `--apply` persists |

---

## FORBIDDEN (hard stop)

Do **NOT**:

1. Run `deeprefine refine` (unless the user explicitly asks for CLI/FAISS mode).

Read the full skill on GitHub

Details

Skill name: deeprefine
Repository: HKUST-KnowComp/DeepRefine-Skill
License: MIT
Repo updated: 23 August 2026

FAQ

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 the Deeprefine skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add HKUST-KnowComp/DeepRefine-Skill --skill deeprefine. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Deeprefine popular?

It lives in HKUST-KnowComp/DeepRefine-Skill, which has 91 GitHub stars and 9 forks.

Related skills

Cachebeat

ARahim3/cachebeat

Keep the Anthropic prompt cache warm by firing a tiny heartbeat only after N minutes of true session inactivity (any exchange resets the clock), so the next real turn reads the conversation at cached price / cached quota instead of a full uncached re-read. The heartbeat is a background Stop hook this skill registers (asyncRewake) - it re-arms itself after every turn, with no tool calls and no permission prompts. Use when the session may sit idle (training runs, CI, downloads, the user stepping away) longer than the cache TTL. Args - idle threshold in minutes (default 50), optional max hours (d

AI, LLMs & Agents10

Stop AI Slop Zh

VincentOld/stop-slop-zh

识别并消除中文 AI 写作的典型痕迹——套话、排比三件套、名词化、抽象主语、金句收尾、总分总八股——让文本读起来像有人在说话，而不是 AI 在生成。适用于公众号、小红书、知乎、公文、学术摘要、产品文案、邮件等所有中文长短文写作场景。

AI, LLMs & Agents

Scammer

naxiaoduo/Scammer.skill

骗子.skill — 把骗局套路蒸馏成 AI，检测可疑消息，定位当前阶段，预判下一步。Distill scam patterns into AI, detect suspicious messages, locate current phase, predict next move.

AI, LLMs & Agents12

Natural Talk

chengzhi-c/natural-talk

自然人味交互与去 AI 腔。消除客服腔、谄媚评判、免责包装、强制平衡与假大空套话；提供专业、平视的同行交流姿态、文本防套路准绳与叙事创作工法（支持日常对话、代码沟通与深度小说创作）；学术公文营销文案不适用。

AI, LLMs & Agents

Turkce Humanizer

bushrabeg/turkce-humanizer

Türkçe metinlerden AI-üretimi yazım imzalarını temizler ve metnin registeri uygunsa insan Türkçesinin ritim imzalarını yerleştirir. "Türkçeleştir," "AI kokusunu at," "bu Türkçe metni doğallaştır," "humanize et" gibi ifadelerle veya AI-üretimi görünen Türkçe paragraf sunulduğunda tetiklenir.

AI, LLMs & Agents

LLM Wiki Manager

sametbrr/llm-wiki-manager

Build, maintain, and query a personal LLM-managed wiki — a markdown knowledge base where the LLM owns all writing, cross-referencing, and bookkeeping while the user curates sources. Use this skill when the user wants to (1) bootstrap a new wiki, second brain, or research notebook; (2) ingest a new source (article, PDF, paper, transcript) and update relevant entity/concept pages, the index, and log; (3) query an existing wiki and synthesize an answer with citations, filing it back if useful; (4) update the wiki when a new source supersedes a claim across multiple pages (diff-before-write sweep)

AI, LLMs & AgentsIn 1 list

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.