Bring your agent

Your agents can drive PromptEden.

Connect Claude Code, Cursor, ChatGPT agents, or your own orchestration over hosted MCP and the v1 API. Reads are scoped. Writes use the same policy, budget, and safety context as Eden.

Use MCP or API keys today. OAuth + PKCE is available for provisioned connector clients. CLI npm packaging is coming soon.

Connect MCP API quickstart

// Remote MCP config

{
  "mcpServers": {
    "prompteden": {
      "url": "https://app.prompteden.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ${PROMPTEDEN_API_KEY}"
      }
    }
  }
}

Agent-first access

without bypassing policy

Hosted MCP

Claude Code, Cursor, ChatGPT agents, and other MCP clients can connect to PromptEden at the hosted MCP endpoint.

https://app.prompteden.com/api/mcp

API keys

Server-side agents can use scoped pe_ keys for REST and MCP bearer-token auth.

Authorization: Bearer pe_...

OAuth + PKCE

Connector clients can use OAuth authorization code with S256 PKCE. Clients are provisioned by PromptEden.

/oauth/authorize -> /oauth/token

CLI

Coming soon

The CLI package is coming soon on npm. Use hosted MCP or API keys today.

@prompteden/cli

REST API v1

Everything agents need, without a parallel trust model.

Wire PromptEden into your own tools with scoped pe_ keys over REST. For the terse machine reference, see the agent API docs.

Accounts, projects, monitors, results, and the provider catalog for the proof layer.
Content topics, articles, and publishing routes for the content engine.
Displacement-scan previews and agent status for orchestration loops.
OAuth connector tokens can call raw REST routes for read-scoped operations; hosted MCP adds Eden's policy and safety layer for writes.

API Request

GET /api/v1/monitors
Authorization: Bearer pe_abc123...

{
  "monitors": [...]
}

Common first calls

route examples

GET /api/v1/accountGET /api/v1/projectsGET /api/v1/monitorsGET /api/v1/monitors/:monitorId/resultsGET /api/v1/content/topicsPOST /api/v1/content/articlesPOST /api/v1/content/articles/:articleId/publishGET /api/v1/agent/status

Policy-aware writes

Agents can do the work. Eden keeps the boundaries intact.

Hosted MCP gives external agents the same scoped identity, policy boundaries, credit ceilings, and safety checks Eden uses in the app.

Read tools return data directly when the bearer token has the required scope.

Hosted MCP resolves the same team, project, policy, and budget context Eden uses in the app.

Routine actions inside those boundaries execute and return an activity receipt to the calling agent.

If a decision genuinely needs human judgment, Eden surfaces it in the conversation with the relevant context.

Agent write path

agent tool call
  → scoped identity
  → policy + budget check
  → execute + receipt
  → ask in Eden only on exception

External agents use Eden's guardrails

one trust model

Same policy system

MCP writes map into Eden's app tools, scopes, budget ceilings, and safety checks instead of creating a parallel trust path.

Taint defense on tool output

Untrusted content is marked, disclosed, or withheld before it can influence a sensitive action.

Scoped identity

API keys and OAuth grants resolve to a team context on the server; agent-supplied tenant fields are not trusted.

Connect the first agent

step-by-step guides