Gleanit
Search and mutate your Gleanit highlights, notes, screenshots, and collections from the agent while researching what to build next.
Overview
co.gleanit/gleanit is an Idea-phase MCP server that lets your agent search and update Gleanit highlights, notes, screenshots, collections, projects, and tags.
What is this MCP server?
- Full Gleanit workspace MCP: search, read, and write highlights, notes, and screenshots
- Manage collections, projects, and tags without leaving the agent session
- Bearer auth via OAuth access token or glean_ API key from Settings → MCP
- Hosted streamable-http remote at api.gleanit.co/api/mcp
- Registry version 1.0.1 for Gleanit-native knowledge operations
- Six entity families exposed: highlights, notes, screenshots, collections, projects, tags
- Two auth methods documented: OAuth bearer and glean_ API key
What problem does it solve?
Research lives in scattered highlights and screenshots that your agent cannot query or update unless you manually paste context every session.
Who is it for?
Solo builders who already use Gleanit as a reading inbox and want agent-driven search, tagging, and collection management.
Skip if: Teams standardized on Notion or Obsidian only, or workflows that do not need write access to a Gleanit account.
What do I get? / Deliverables
After connecting Gleanit MCP, your agent can read and write your knowledge base so ideas stay linked to sources as you narrow what to build.
- Agent-searchable Gleanit corpus with updated notes and tags
- Structured collections and projects maintained via tool calls
- Less copy-paste of highlights into coding sessions
Recommended MCP Servers
Journey fit
Research artifacts accumulate earliest in Idea when you capture quotes, competitors, and inspiration before scope lock. Research is the shelf for reading, clipping, and reorganizing external knowledge into projects and tags.
How it compares
Personal knowledge MCP for Gleanit, not a public web research or SEO crawler skill.
Common Questions / FAQ
Who is co.gleanit/gleanit for?
Indie researchers and founders who capture web highlights in Gleanit and want MCP read/write from Claude Code, Cursor, or similar clients.
When should I use co.gleanit/gleanit?
Use it in Idea and early Validate when you are synthesizing competitors, pricing, and user evidence stored as Gleanit notes and screenshots.
How do I add co.gleanit/gleanit to my agent?
Generate OAuth or a glean_ API key under Gleanit Settings → MCP, then add https://api.gleanit.co/api/mcp with Authorization Bearer in your MCP config.