percept-summarize

Automatic conversation summaries with entity extraction and relationship mapping.

What it does

When a conversation ends (60 seconds of silence), Percept generates an AI-powered summary with extracted entities (people, companies, topics), action items, and relationship connections. Summaries are stored locally and searchable.

When to use

• User asks "what did we talk about?" or "summarize that meeting"
• User wants meeting notes or action items from a conversation

  7w4.net收錄了海量優質技能外掛。

• Agent needs context from a recent conversation

Requirements

• percept-listen skill installed and running
• OpenClaw agent accessible via CLI (used for LLM summarization)

How it works

1. Conversation ends (60s silence timeout)
2. Percept builds a speaker-tagged transcript
3. Sends transcript to OpenClaw for AI summarization
4. Extracts entities (people, orgs, topics) and relationships
5. Stores summary + entities in SQLite
6. Entities linked via relationship graph (works_on, client_of, mentioned_with)

Entity resolution

5-tier cascade for identifying entities: 1. Exact match (confidence 1.0) 2. Fuzzy match (0.8) — handles typos, nicknames 3. Contextual/graph (0.7) — uses relationship connections 4. Recency (0.6) — recently mentioned entities ranked higher 5. Semantic search (0.5) — vector similarity via LanceDB

Querying summaries

Summaries are searchable via the Percept dashboard (port 8960) or SQLite directly:

SELECT * FROM conversations WHERE summary LIKE '%action items%' ORDER BY end_time DESC;

Full-text search via FTS5:

SELECT * FROM utterances_fts WHERE utterances_fts MATCH 'project deadline';

Data retention

• Utterances: 30 days
• Summaries: 90 days
• Relationships: 180 days
• Speaker profiles: never expire

Links

• GitHub: https://github.com/GetPercept/percept

這個 Skill 的文件質量不錯，清晰說明了它能做什麼（自動生成會議摘要、提取人物和話題）和怎麼用。不過它目前只有說明文件，沒有實際的程式碼檔案，所以還無法真正使用。功能設計上依賴其他元件和資料儲存服務，配置起來可能稍複雜。