Make your AI agentsaccountable.
One schedule for all your agents: Set it. Watch it run. Get proof it worked.
# Your agent. On time. Every time.curl -X POST https://api.cueapi.ai/v1/cues \-H "Authorization: Bearer cue_sk_..." \-H "Content-Type: application/json" \-d '{"name":"morning-briefing","schedule":{"type":"recurring","cron":"0 9 * * *"},"callback":{"url":"https://your.agent/run"}}'# HTTP 200{ "id": "cue_a3f9b2...", "status": "active" }
Schedule. Deliver. Confirm.
The problem
What cron doesn’t tell you
Silent failures
Cron fails. No alert. No log. Your agent missed its window and nobody knows until a customer complains.
No proof of execution
Your agent says it ran. Did it? Cron gives you nothing. No receipt. No timestamp. No outcome. Just trust.
Cron was built in 1975
No retries. No observability. No API. You're running 2026 agents on 1975 infrastructure.
Three steps. Full accountability.
Schedule → Deliver → Confirm
Schedule
Register a cue with a schedule and callback URL
$ curl -X POST https://api.cueapi.ai/v1/cues \
-H "Authorization: Bearer cue_sk_..." \
-d '{
"name": "morning-brief",
"schedule": "0 9 * * *",
"callback_url": "https://my-agent.com/hook",
"payload": { "task": "brief" }
}'
{
"id": "cue_a1b2c3d4e5f6",
"status": "active",
"next_fire_at": "2025-01-15T09:00:00Z"
}Deliver
CueAPI fires your webhook on time. Retries on failure.
POST https://my-agent.com/hook
X-CueAPI-Signature: sha256=9f3a...
Content-Type: application/json
{
"execution_id": "exec_7g8h9i",
"cue_id": "cue_a1b2c3d4e5f6",
"fired_at": "2025-01-15T09:00:00Z",
"payload": { "task": "brief" }
}
→ 200 OKConfirm
Your handler reports the outcome. CueAPI tracks everything.
$ curl -X POST https://api.cueapi.ai/v1/executions/exec_7g8h9i/outcome \
-H "Authorization: Bearer cue_sk_..." \
-d '{
"status": "success",
"detail": "Brief sent to #general"
}'
{
"execution_id": "exec_7g8h9i",
"outcome": "success",
"recorded_at": "2025-01-15T09:00:03Z"
}Live execution feed
Simulated CueAPI executions — every 2.5 seconds
| Time | Cue Name | Delivery | Outcome | Duration |
|---|
Built for reliability
Everything your agents need to run on schedule
Cron vs CueAPI
What you get when scheduling is more than a one-liner
| Feature | Cron | CueAPI |
|---|---|---|
| Scheduling | Cron expression on your server | Managed cron via API — no infra needed |
| Delivery confirmation | None — fire and forget | HTTP-level delivery receipt on every execution |
| Outcome tracking | None — no idea if the job succeeded | Structured outcome reports with verification |
| Retries | None — you build retry logic yourself | Automatic exponential backoff, up to 5 attempts |
| Failure notification | None unless you build alerting | Built-in alerts via webhook on any failure |
| Execution history | None — check your own logs | Full audit trail with API and dashboard access |
Start in minutes
One API call to schedule your first cue
curl -X POST https://api.cueapi.ai/v1/cues \-H "Authorization: Bearer cue_sk_..." \-H "Content-Type: application/json" \-d '{"name": "morning-briefing","schedule": {"type": "recurring", "cron": "0 9 * * *"},"callback": {"url": "https://my-agent.com/webhook"},"payload": {"task": "generate-briefing"}}'
12 agent scheduling patterns
From daily briefings to multi-agent orchestration
Morning Briefing
Daily summary delivered at 9am
Email Digest
Process and summarize overnight emails
Follow-up Sequences
Automated follow-up at the right time
Churn Detection
Monitor customer health signals daily
System Monitoring
Infrastructure health checks every 5 min
Database Backup
Verified backup completion with proof
Agent Coordinator
Orchestrate multiple agents on schedule
Content Pipeline
Generate, review, and publish content
Data Scraping
Scheduled data collection with retry
Code Review
Automated PR review on every push
Invoice Processing
Process invoices on the 1st of each month
Weekly Report
Executive summary every Friday at 5pm
Simple, transparent pricing
Every plan includes every feature. Scale when you need to.
Free
10 cues
300 executions/mo
60 req/min
Pro
100 cues
5,000 executions/mo
200 req/min
Scale
500 cues
50,000 executions/mo
500 req/min
Everything from the command line
pip install cueapi — manage cues without leaving the terminal
Self-host the entire engine
The core scheduling engine is open source. Run it anywhere with Docker Compose.
$ docker compose up -d
# CueAPI running at http://localhost:8000Security built in, not bolted on
Enterprise-grade security on every plan
Signed Webhooks
HMAC-SHA256 signatures on every delivery
Hashed API Keys
SHA-256 hashed. Plaintext never stored.
SSRF Protection
Callback URLs validated. Private IPs blocked.
Replay Protection
Timestamp-bound signatures. 5-minute window.
HTTPS Only
Production requires HTTPS callback URLs.
Audit Trail
Every execution logged with delivery + outcome.
No public URL? No problem.
Worker transport pulls jobs from CueAPI. Runs on Mac Mini, Replit, private servers. Anywhere.
Install
pip install cueapi-workerConfigure
# cueapi-worker.yaml
handlers:
morning-brief:
command: python brief.py
health-check:
command: ./check.shStart
cueapi-worker start -c cueapi-worker.yaml$ pip install cueapi-worker
Successfully installed cueapi-worker-0.4.1
$ cueapi-worker start -c cueapi-worker.yaml
[worker] Connected to wss://api.cueapi.ai/ws
[worker] Registered handlers: morning-brief, health-check
[worker] Polling for jobs...
[worker] Received job exec_7g8h9i → morning-brief
[worker] Running: python brief.py
[worker] Reported outcome: success
Schedule it once. Trust it forever.
Let the agent be the brain. Let CueAPI be the clock.