Charla for AI agents
Charla combines live chat and an AI-assisted knowledge base, letting a support team manage conversations and help articles from one place. 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 Charla.
Look up a contact's history
Look up a contact's details before jumping into an ongoing support conversation.
Publish a knowledge base article
Create a new knowledge base article covering a common, recurring support question.
Confirm before deleting an article
Confirm before deleting a knowledge base article, since customers may still be linking to it.
9 Charla actions, graded by risk.
Every Charla action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
6Look things up. Allowed by default.
- charla.get_contactGet Contact
- charla.list_contactsList Contacts
- charla.get_kb_articleGet Knowledge Base Article
- charla.list_kb_articlesList Knowledge Base Articles
- charla.list_conversationsList Conversations
- charla.list_kb_categoriesList Knowledge Base Categories
write
2Create and change things. Allow, or ask the user first.
- charla.create_kb_articleCreate Knowledge Base Article
- charla.update_kb_articleUpdate Knowledge Base Article
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- charla.delete_kb_articleDelete Knowledge Base Article
Charla in three steps.
- 01Your users connect CharlaThey add their Charla api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Knowledge Base Article” can wait for the user, and “delete Knowledge Base Article” can be denied outright.
- 03Any agent can actYour agent calls Charla through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('charla', { read: 'allow', write: 'ask', // create_kb_article destructive: 'deny', // delete_kb_article }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Charla connects.
Users add their Charla 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 Charla 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 Charla from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Charla and Arc0, answered.
Can I use Charla with Claude, ChatGPT or Cursor?
Yes. Connect Charla 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 Charla actions you allow.
How do users connect Charla?
Users add their Charla 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 Charla actions can my agent take?
9 in total: 6 read, 2 write and 1 destructive, such as “create Knowledge Base Article”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Charla?
Yes. Actions like “delete Knowledge Base Article” 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 Charla too?
Yes. The same Charla connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Charla into your agent.
Your users connect Charla 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