canonical: https://jentic.com/apis/agenty.com/agenty-main

# Agenty API

Jentic publishes the only available OpenAPI specification for Agenty API, keeping it validated and agent-ready. This is the curated 21-endpoint cut of the Agenty platform covering the core surface most integrations need: agents, jobs, lists, scheduler, workflows, users, connections, API keys, dashboard, and projects. The platform itself does the web data extraction - capture screenshots, track redirects, and convert URLs into LLM-ready markdown - while this API drives execution and reads results.

## For AI agents

Run pre-built Agenty agents to extract structured web data, capture screenshots, or generate LLM-ready markdown; manage lists, schedules, and workflows.

## Scope

Does not author scraping logic, render HTML, manage proxies, or solve CAPTCHAs - use for executing pre-built Agenty agents and workflows only.

## Capabilities

- Start a job on an existing Agenty agent and retrieve the result via /jobs and `/jobs/{id}/result`
- Convert URLs into LLM-ready markdown by running an Agenty agent configured for that output
- Maintain reusable URL lists that drive batch scrapes via /lists and `/lists/{id}/rows`
- Schedule recurring runs of agents and workflows via /scheduler
- Chain multiple agents into a sequenced workflow via /workflows

## Use cases

### Programmatic Web Scraping

Run scrapers built in the Agenty UI from external systems by calling POST /jobs to start a job and then GET `/jobs/{id}/result` for parsed output. Agenty handles browser orchestration, retries, and storage. Suited to teams who prefer a visual scraper builder but need programmatic execution.

Example prompt: POST /jobs with agent_id=42 and input list_id=11, poll until completed, then fetch the results

### URL-to-Markdown for LLM Pipelines

Use an Agenty agent configured to output Markdown to convert a list of URLs into LLM-ready text for retrieval or summarisation. The agent handles JavaScript rendering and content extraction; the calling code only manages the input list and result download. Useful for building a custom RAG corpus from public web pages.

Example prompt: Update list 7 with 50 new URLs, start the markdown-conversion agent, and upload the resulting markdown chunks to a vector store

### Scheduled Monitoring

Configure /scheduler entries to run an agent or workflow at fixed times so price changes, availability, or content updates are captured automatically. The agent reads back the latest job result on each schedule tick. Suitable for competitive monitoring or stock trackers.

Example prompt: Create a daily scheduler entry at 06:00 UTC for workflow 9 and email the result diff to a Slack channel

### Agent-Driven Web Research

Let a Jentic-connected agent select an Agenty agent, kick off a job, and reason over the results inside a longer plan. The agent searches Jentic for the start-job operation, loads its schema, and chains it with `/jobs/{id}/result.` This makes Agenty a callable scraping primitive for autonomous research.

Example prompt: Given a question, choose the appropriate Agenty agent, start a job with the relevant URLs, and summarise the results

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | `/jobs` | Start a new job |
| GET | `/jobs/{job_id}` | Get job status |
| GET | `/jobs/{job_id}/result` | Fetch job results |
| GET | `/agents` | List agents |
| POST | `/lists/{list_id}/rows` | Append rows to an input list |

## Key resources

- **Agents** — Pre-built scraping or automation definitions executed via the API
- **Jobs** — Single runs of an agent with status, results, and logs
- **Lists** — URL lists used as inputs for batch runs
- **Scheduler** — Recurring schedules attached to agents or workflows
- **Workflows** — Sequenced chains of agents with conditional execution

## Why Jentic

- **Setup:** Wiring the Agenty API by hand means managing its bearer auth and coding your own polling across job, agent, and list operations. Through Jentic you install once, import the Agenty API from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** Agenty puts the job id and list id in the URL path (`/jobs/{job_id}`, `/lists/{list_id}/rows`), so a rule can pin the agent to one job or list. You choose which operations it may call, such as reading a job result, so job creation is not included unless you add it.
- **Credential handling:** Your Agenty bearer token is stored once, encrypted, by your own Jentic One instance and injected as the Authorization header at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'start a scraping job' or 'convert a URL to markdown', and Jentic returns the matching Agenty operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Apify API** — Scraping marketplace with maintained Actors for major sites and dataset storage
- **Bannerbear API** — Image and screenshot generation API focused on templated visuals
- **ZenRows** — Proxy and anti-bot bypass for hard-to-scrape targets

## FAQ

### Why is there no official OpenAPI spec for Agenty API?

Agenty does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Agenty API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the Agenty API use?

Agenty uses HTTP bearer token authentication. Pass the token in the Authorization header as 'Bearer <token>'. Through Jentic the token is held in the credential vault and injected at execution time.

### Can I convert a URL to LLM-ready markdown with the Agenty API?

Yes, via an Agenty agent configured for markdown output. POST /jobs with that agent_id and a single-URL input; the resulting job result contains the markdown payload. The Agenty platform handles JavaScript rendering and content extraction.

### What are the rate limits for the Agenty API?

Limits are tied to your plan's concurrent-job and API-call quotas rather than a single requests-per-second cap. Watch for 429 responses, retry with backoff, and respect the concurrent-job ceiling when fanning out runs.

### How do I run an Agenty agent through Jentic?

Use the Jentic search 'start an agenty job'. Jentic returns POST /jobs; load its schema, supply the agent_id and any input list_id, and execute. The response contains the job_id which you poll via GET `/jobs/{job_id}.`

### Does Agenty support workflow chaining via the API?

Yes. The /workflows endpoints let you create and execute chains of agents with sequential and conditional steps, so multi-stage scrapes can be defined and triggered from this API.

### Can I limit what my agent is allowed to do with the Agenty API?

Yes. Because you run Jentic One yourself, your own rules decide which Agenty operations and credentials the agent may use. Agenty puts the job id and list id in the URL path (`/jobs/{job_id}`, `/lists/{list_id}/rows`), so a rule can pin the agent to a single job or list. You can also allow only specific operations, such as reading a job result via GET `/jobs/{job_id}/result` while withholding job creation on POST /jobs.