$ loading_
Compress local text context for AI agents with retrieval and stats.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "winnow" yet — see the docs or source repo.
Use winnow to compress this 20,000-word product requirements document. Keep goals, core features, constraints, and open questions, then return the compressed result and compression statistics.
A condensed context ready for AI input, plus statistics such as compression ratio.
I have meeting notes compressed with winnow. Help me locate the summary segment about budget approval and retrieve the corresponding original text.
The relevant compressed segment is identified, and its original source text is returned.
Run winnow compression on this batch of customer support conversations and summarize the compression statistics for each file to evaluate which texts are best suited for context compression.
A summary of compression statistics for each text and an initial assessment of suitability for compression-based workflows.
Compresses LLM conversation context while preserving meaning and reducing token usage.
Compress Claude Code and OpenCode context losslessly for longer, more efficient coding tasks.
Detect AI-generated content, check plagiarism, analyze images, and compare texts.
Automate Winston AI workflows through Rube MCP for detection and result handling.
Query multiple AI models in parallel and synthesize more balanced answers.
Monitor, audit, rewind, and safely test every AI agent action.