API Reference

Base URL: https://api.continua.ai/personal/v1

All requests require Authorization: Bearer cai_YOUR_KEY unless noted otherwise.

Memory

List claims

GET /memory/claims
GET /memory/claims?category=preference

Returns active memory claims. Filter by category: preference, habit, relationship, opinion, fact.

Search claims

GET /memory/claims/search?q=coffee

Semantic search across all claims.

Personality profile

GET /memory/claims/profile

Derived personality model from the user's full memory.

Update a claim

PUT /memory/claims/{id}
Content-Type: application/json

{"content": "Prefers oat milk", "reason": "user corrected"}

Delete a claim

DELETE /memory/claims/{id}

Change visibility

PUT /memory/claims/{id}/scope
Content-Type: application/json

{"scope": "contacts"}   # private | contacts | public

Pin / unpin

PUT /memory/claims/{id}/pin
Content-Type: application/json

{"pinned": true}

Revision history

GET /memory/claims/{id}/history

Agent

Run agent

POST /agent/run
Content-Type: application/json

{"message": "What should I have for dinner?"}

Add Accept: text/event-stream for SSE streaming.

Response (sync):

{"content": "Based on what you've told me...", "iterations": "3", "duration_ms": "1523"}

Send P2P message

POST /agent/message
Content-Type: application/json

{"to": "alice@example.com", "content": "Share your travel tips?"}

List threads

GET /agent/threads

Thread messages

GET /agent/threads/{id}/messages

Flux

Create app

POST /flux/create
Content-Type: application/json

{"prompt": "Team lunch poll: pizza, sushi, or tacos?"}

Response:

{
  "app_id": "flux-abc123",
  "app_url": "https://flux.continua.ai/flux-abc123",
  "embed": "<iframe src=\"...\" width=\"100%\" height=\"600\"></iframe>"
}

Interact

POST /flux/{app_id}/interact
Content-Type: application/json

{"intent": "vote", "data": {"choice": "sushi"}}

Get state

GET /flux/{app_id}/state

WebSocket (real-time)

const ws = new WebSocket(
  `wss://api.continua.ai/personal/v1/flux/${appId}/ws?token=${apiKey}`
);
ws.onmessage = (e) => console.log(JSON.parse(e.data));

Feeds

Create feed

POST /feeds
Content-Type: application/json

{
  "name": "Downtown Café Status",
  "description": "Real-time wait times",
  "type": "status",
  "tags": ["café", "downtown"],
  "cost_per_update": 2,
  "visibility": "public",
  "location": {"lat": 37.7749, "lng": -122.4194, "radius_m": 500, "zone": "SF"}
}

Types: presence · inventory · status · interests · household · commerce · community · health · profession

Visibility: public · contacts · link · private

List my feeds

GET /feeds

Get feed

GET /feeds/{id}

Delete feed

DELETE /feeds/{id}

Publish update

POST /feeds/{id}/publish
Content-Type: application/json

{
  "data": {"tables_available": "5", "wait_minutes": "0"},
  "summary": "5 tables available, no wait"
}

Discover feeds

GET /feeds/discover?type=status
GET /feeds/discover?tag=café
GET /feeds/discover?q=coffee
GET /feeds/discover?lat=37.77&lng=-122.42&radius_m=2000

POST /feeds/{id}/subscribe
Content-Type: application/json

{"budget": 100}   # max credits/month, 0 = unlimited

Unsubscribe

DELETE /feeds/{id}/subscribe

List subscriptions

GET /feeds/subscriptions

Stream (SSE)

GET /feeds/{id}/stream
Accept: text/event-stream

Requires active subscription. Sends event: latest (initial) then event: update for each new item.

Stream (WebSocket)

const ws = new WebSocket(
  `wss://api.continua.ai/personal/v1/feeds/${feedId}/ws?token=${apiKey}`
);

Publisher analytics

GET /feeds/analytics/overview    # dashboard summary
GET /feeds/{id}/analytics        # per-feed detail
GET /feeds/analytics/top         # ranked by revenue
GET /feeds/earnings              # current period (or ?period=2026-01-01)

Channels

Create channel

POST /channels
Content-Type: application/json

{
  "name": "Downtown Food Scene",
  "description": "Best food feeds in downtown",
  "feed_ids": ["feed_abc", "feed_def"],
  "markup_pct": 15
}

List channels

GET /channels

Get channel

GET /channels/{id}

Subscribe to channel

POST /channels/{id}/subscribe

Skills

GET /skills                           # list skills
GET /skills/{name}                    # get skill details

Service Catalog & Connections

The service catalog contains 61+ web services the AI can connect to. Connections store credentials and auto-generate skills so the agent can act on the user's behalf.

Browse catalog

GET /catalog                          # all entries (builtin + user-added)
GET /catalog/search?q=calendar        # search by name/tags

Each entry includes: id, display_name, auth_method (oauth/api_key/cookie/linking_code/none), category, tags, value_prop, and verified status.

Connect a service

Requires scope: catalog:write

# API key auth
curl -X POST https://api.continua.ai/personal/v1/catalog/connect \
  -H "Authorization: Bearer $API_KEY" \
  -d '{"entry_id": "openai", "api_key": "sk-..."}'

# Linking code auth
curl -X POST https://api.continua.ai/personal/v1/catalog/connect \
  -d '{"entry_id": "spotify", "linking_code": "123456"}'

# OAuth (returns redirect URL)
curl -X POST https://api.continua.ai/personal/v1/catalog/connect \
  -d '{"entry_id": "google-calendar"}'
# Response: {"status": "oauth_redirect", "oauth_url": "https://..."}

Response includes status (connected/oauth_redirect), skill_name, skill (SKILL.md content), auth_flow, and credential_stored.

List connections with health status

GET /catalog/connections

Returns all services with connection status: connected, needs_refresh, failed, pending, or not_connected. Includes hint for fixing broken connections.

Verify connection health

POST /catalog/connections/{service}/verify

Probes the service API with stored credentials. Returns detailed status including http_status and actionable hint.

Disconnect