# Chat Refiner

> Refine conversation history by deleting useless/redundant/process exchanges. Keep explicit instructions, disciplines, important configs, skills learned (how), user 'remember this' info. Use when user asks to summarize/clean chat logs, sessions_history, or memory maintenance. Produces concise MEMORY.md updates or summary files.

- Skill: `modbender/chat-refiner` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add modbender/chat-refiner`
- Raw SKILL.md: https://api.skillmd.com/api/skills/modbender/chat-refiner/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: modbender (https://skillmd.com/u/modbender)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/modbender/chat-refiner

---


# Chat Refiner

## When to Use

User requests:
- "精简聊天记录"
- "clean/summarize history"
- "update MEMORY.md from transcripts"
- Memory maintenance during heartbeats.

## Workflow

1. **Input**: sessions_history (this session or other), memory/*.md, chat transcripts.
2. **Principles** (see references/principles.md):
   - Delete casual/heartbeats/repeated tools.
   - Keep: disciplines, configs, skills learned/install, explicit "remember", decisions.
3. **Process**:
   - Read raw.
   - Extract key.
   - Write summary to memory/YYYY-MM-DD-summary.md or MEMORY.md.
4. **Output**: Clean file + "Refined X items kept".

## Tools

- sessions_history: fetch transcripts.
- memory_search/get: prior context.
- write/edit: output summary.

## Examples

User: "精简对话" → read history → refine → write memory/summary.md.

Ref: [principles.md](references/principles.md)