Glasp for AI agents
Glasp saves bookmarks and highlights from articles, books, and Kindle, and generates AI summaries of them. Readers and researchers use it to build a personal knowledge archive. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Glasp.
Save a bookmark for later
Save an article URL as a bookmark so it is captured in a reading list for later review.
Pull highlights from a source
List the highlights saved from a specific article or book to reuse them in a summary.
Fetch an AI-generated summary
Retrieve the existing summary for a saved item instead of re-reading the full source.
9 Glasp actions, graded by risk.
Every Glasp action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
5Look things up. Allowed by default.
- glasp.get_summaryGet Summary
- glasp.list_bookmarksList Bookmarks
- glasp.list_summariesList Summaries
- glasp.list_highlightsList Highlights
- glasp.list_kindle_highlightsList Kindle Highlights
write
3Create and change things. Allow, or ask the user first.
- glasp.update_highlightUpdate Highlight
- glasp.create_highlightsCreate Highlights
- glasp.save_bookmarkSave Bookmark
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- glasp.delete_highlightsDelete Highlights
Glasp in three steps.
- 01Your users connect GlaspThey add their Glasp api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Highlights” can wait for the user, and “delete Highlights” can be denied outright.
- 03Any agent can actYour agent calls Glasp through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('glasp', { read: 'allow', write: 'ask', // create_highlights destructive: 'deny', // delete_highlights }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Glasp connects.
Users add their Glasp api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Glasp connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Glasp from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Glasp and Arc0, answered.
Can I use Glasp with Claude, ChatGPT or Cursor?
Yes. Connect Glasp to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Glasp actions you allow.
How do users connect Glasp?
Users add their Glasp api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Glasp actions can my agent take?
9 in total: 5 read, 3 write and 1 destructive, such as “create Highlights”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Glasp?
Yes. Actions like “delete Highlights” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Glasp too?
Yes. The same Glasp connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Glasp into your agent.
Your users connect Glasp once, under your brand. Your agent gets 9 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan