5.4 KiB
Chat Summarization
Chat Summarization allows you to compress long conversations into concise AI-generated summaries. This helps save context tokens and makes it easier to continue work on complex topics without losing important information.
Overview
When conversations grow long, they consume more context tokens with each message. Chat Summarization uses your configured Chat Assistant provider to create an intelligent summary that preserves:
- Key decisions and conclusions
- Technical details and code references
- Important context for continuing the conversation
Key Features:
- One-click compression: Summarize directly from the chat bottom bar
- Preserves original: Creates a new chat file, keeping the original intact
- Smart summaries: AI extracts the most relevant information
- Markdown formatted: Summaries are well-structured and readable
Using Chat Summarization
Compressing a Chat
- Open any chat with conversation history
- Click the Compress button (📦) in the chat bottom bar
- Wait for the AI to generate the summary
- A new chat opens with the compressed summary
Agent Sessions: Hand Over
In a chat bound to an ACP agent the same button reads Hand over. Instead of writing a compressed chat file, it summarizes the transcript using your configured Chat Assistant provider (the same LLM that powers compression) and starts a fresh agent session with the summary as its opening context. The transcript in the chat stays intact; only the agent-side session is replaced.
The button is disabled with an explanatory tooltip when the Chat Assistant configuration is incomplete (provider, model, template, or URL missing), because the summary is generated by that configuration. Auto-compression never triggers for agent sessions — the agent manages its own context window.
What Gets Preserved
The summarization process:
- Maintains chronological flow of the discussion
- Keeps technical details, code snippets, and file references
- Preserves key decisions and conclusions
- Aims for 30-40% of the original conversation length
What Gets Filtered
The following message types are excluded from summarization:
- Tool call results (file reads, searches)
- File edit blocks
- Thinking/reasoning blocks
How It Works
┌─────────────────────────────────────────────────────────────┐
│ CHAT SUMMARIZATION │
├─────────────────────────────────────────────────────────────┤
│ 1. Original chat messages are collected │
│ 2. Tool/thinking messages are filtered out │
│ 3. AI generates a structured summary │
│ 4. New chat file is created with summary as first message │
│ 5. Original chat remains unchanged │
└─────────────────────────────────────────────────────────────┘
File Naming
Compressed chats are saved with a unique suffix:
original_chat.json → original_chat_a1b2c.json
Both files appear in your chat history, allowing you to switch between the full conversation and the summary.
Best Practices
- Summarize at natural breakpoints: Compress after completing a major task or topic
- Review the summary: Ensure important details were captured before continuing
- Keep originals: Don't delete original chats until you've verified the summary is sufficient
- Use for long sessions: Most beneficial for conversations with 20+ messages
When to Use
Good candidates for summarization:
- Long debugging sessions with resolved issues
- Feature implementation discussions with final decisions
- Research conversations where conclusions were reached
- Any chat approaching context limits
Consider keeping full history for:
- Ongoing work that may need exact message references
- Conversations with important code snippets you'll copy
- Discussions where the reasoning process matters
Configuration
Chat Summarization uses your current Chat Assistant settings:
- Provider: Same as Chat Assistant (Settings → QodeAssist → General)
- Model: Same as Chat Assistant
- Template: Same as Chat Assistant
No additional configuration is required.
Troubleshooting
Compression Button Disabled
- Hover the button: the tooltip explains why (for example, the Chat Assistant configuration is unassigned, or an agent chat has nothing to summarize yet)
- Assign the chat feature in Preferences → QodeAssist if the tooltip says so
- While a compression or summary is running, the button is replaced by a progress indicator with a Cancel action in the bottom bar
Compression Fails
- Verify your Chat Assistant provider is configured correctly
- Check network connectivity
- Ensure the model supports chat completions
Summary Missing Details
- The AI aims for 30-40% compression; some details may be condensed
- For critical information, keep the original chat
- Consider summarizing smaller conversation segments
Related Documentation
- File Context - Attach files to chat