# Show HN: Cronloop, run Claude Code or Codex on a schedule

> Source: <https://cronloop.ai/>
> Published: 2026-08-18 16:31:41+00:00

[NewManage your agents from ChatGPT and Claude](/mcp)

# AI agents that run in a loop.

Cronloop keeps AI agents running for you, from every five minutes to once a week. Describe the job in plain Markdown, pick Codex or Claude Code, then watch each run live. Every agent keeps a memory between runs.

CodexClaude CodeRuns with your subscriptions or provider API keys.

### 1Give your agent instructions

## Create specialized agents in minutes

An agent is whatever you describe in Markdown. You can create agents for sales, support, growth, engineering, and more.

### Book outbound demos

“Book qualified demos in our ICP with high-touch email from its own inbox, every 30 minutes.”

### Resolve support tickets

“Respond to support tickets by reading our support docs, replying over email, and escalating anything sensitive to a human, every 5 minutes.”

### Grow organic traffic

“Lift search rankings by reading fresh Search Console data and shipping measured content updates to our CMS, every morning.”

### Turn issues into pull requests

“Open pull requests by picking up the next assigned issue and writing the fix and the tests, every hour.”

### Automate social content

“Grow our social following by turning our content calendar into on-brand posts for every channel, three times a day.”

### Track your competitors

“Watch our competitors by reading their changelogs, pricing pages, and launches, and posting a digest to Notion, twice a day.”

If you can write the job description, you can run the agent.

## Any tool. Any API. Any skill.

Every run starts in a fresh sandbox where your agent can install whatever the job needs. Connect the software your team already uses, or add a skill from the catalog.

If it has an API, MCP, or CLI, your agent can use it.

## Run Cronloop from ChatGPT or Claude

Your agents are also reachable from the assistant you already talk to. Add one address, approve it once, and ask for a new agent, a fresh run, or a post-mortem on last night's failures without opening a dashboard.

- 1
### Add the address

Open the connector settings in Claude, ChatGPT, or any other MCP client and add Cronloop as a custom connector.

- 2
### Approve the connection

Sign in to Cronloop and approve access. It is a standard OAuth sign-in, so there is no key to copy, paste, or keep safe.

- 3
### Ask for what you need

Your assistant now speaks to Cronloop directly. Ask it in plain language and it does the work in your account.

Cronloop's address

`https://app.cronloop.ai/v1/mcp`

Works with any MCP client that supports remote servers, including Claude and ChatGPT.

Set up an agent that answers new Zendesk tickets from our help docs every 5 minutes and escalates anything sensitive to me.

`agents_create`

Support triage, every 5 minutes`secrets_set`

Zendesk token stored, write-only`runs_start`

First run started`runs_get`

Answered 6 tickets, escalated 1

Support triage is live. Its first run answered 6 tickets from your help docs and escalated one refund request to you. It will keep running every 5 minutes.

### Agents

Create an agent, rewrite its instructions, change its schedule, then pause, resume, or delete it.

`agents_list`

`agents_get`

`agents_create`

`agents_update`

`agents_pause`

`agents_resume`

`agents_delete`

### Runs

Start a run on demand, follow it to the end, cancel it, and review what recent runs actually did.

`runs_start`

`runs_list`

`runs_review`

`runs_get`

`runs_cancel`

### Settings and secrets

Add the settings and credentials a job needs. Values are encrypted on the way in and can never be read back out.

`configuration_list`

`environment_set`

`environment_delete`

`secrets_set`

`secrets_delete`

### Memory

Read, write, and clear the durable memory each agent carries from one run to the next.

`memory_list`

`memory_read`

`memory_write`

`memory_delete`

## Pricing

Use your existing ChatGPT or Claude subscription, or bring an Anthropic or OpenAI API key. Model usage stays with your chosen provider connection.

Works with provider subscriptions and API keys

### Free

Put your first agents to work.

$0per month

- Up to 3 agents
- Runs as often as every hour
- 200 run minutes each month
- Runs up to 10 minutes
- 1 active run at a time
- Provider subscriptions or API keys
- Live output and run history

### Pro

For work that cannot wait an hour.

$25per month

$20 per month when billed annually.

- No fixed agent limit
- Runs as often as every 5 minutes
- No fixed monthly run-minute limit
- Runs up to 60 minutes
- 3 active runs at a time
- Provider subscriptions or API keys
- Priority support

## Give an agent its first job.

Sign in with GitHub, Google, or your work email and put your first agent to work in a few minutes.
