Dovetail for AI agents
Dovetail is a research analysis platform that organizes interview transcripts, survey responses, and customer feedback into tagged, summarized insights for product and research teams. 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 Dovetail.
Pull a project's insights
Get the insights saved under a research project to summarize findings for a stakeholder update.
Tag a new note
Create a note and topic tag for a fresh piece of customer feedback so it is searchable later.
Confirm before deleting an insight
Require approval before deleting an insight, since it may already be cited in a shared report.
51 Dovetail actions, graded by risk.
Every Dovetail action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
24Look things up. Allowed by default.
- dovetail.get_docGet Doc
- dovetail.get_dataGet Data
- dovetail.get_fileGet File
- dovetail.get_noteGet Note
- dovetail.list_dataList Data
- dovetail.list_docsList Docs
- dovetail.list_tagsList Tags
- dovetail.get_folderGet Folder
- dovetail.list_notesList Notes
- dovetail.get_contactGet Contact
- dovetail.get_insightGet Insight
- dovetail.get_projectGet Project
- dovetail.list_foldersList Folders
- dovetail.list_contactsList Contacts
- dovetail.list_insightsList Insights
- dovetail.list_projectsList Projects
write
21Create and change things. Allow, or ask the user first.
- dovetail.create_docCreate Doc
- dovetail.update_docUpdate Doc
- dovetail.create_dataCreate Data
- dovetail.create_noteCreate Note
- dovetail.update_dataUpdate Data
- dovetail.update_noteUpdate Note
- dovetail.create_topicCreate Topic
- dovetail.update_topicUpdate Topic
- dovetail.create_channelCreate Channel
- dovetail.create_contactCreate Contact
- dovetail.create_insightCreate Insight
- dovetail.create_projectCreate Project
- dovetail.update_channelUpdate Channel
- dovetail.update_contactUpdate Contact
- dovetail.update_insightUpdate Insight
- dovetail.create_data_pointCreate Data Point
destructive
6Delete, cancel or archive. Ask first, or deny outright.
- dovetail.delete_docDelete Doc
- dovetail.delete_dataDelete Data
- dovetail.delete_noteDelete Note
- dovetail.delete_topicDelete Topic
- dovetail.delete_channelDelete Channel
- dovetail.delete_insightDelete Insight
Dovetail in three steps.
- 01Your users connect DovetailThey add their Dovetail api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Doc” can wait for the user, and “delete Doc” can be denied outright.
- 03Any agent can actYour agent calls Dovetail through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('dovetail', { read: 'allow', write: 'ask', // create_doc destructive: 'deny', // delete_doc }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Dovetail connects.
Users add their Dovetail 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 Dovetail 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 Dovetail from any agent.
Dovetail and Arc0, answered.
Can I use Dovetail with Claude, ChatGPT or Cursor?
Yes. Connect Dovetail 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 Dovetail actions you allow.
How do users connect Dovetail?
Users add their Dovetail 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 Dovetail actions can my agent take?
51 in total: 24 read, 21 write and 6 destructive, such as “create Doc”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Dovetail?
Yes. Actions like “delete Doc” 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 Dovetail too?
Yes. The same Dovetail connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Dovetail into your agent.
Your users connect Dovetail once, under your brand. Your agent gets 51 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan