# Gumloop (slim)

> Gumloop is a hosted platform from AgentHub Inc. for building AI agents that work across company apps and run on schedules, webhooks and app events. It has a REST API, a Python SDK, a CLI and a hosted MCP server.

- Full: https://www.anchorterminal.com/tools/gumloop.md (~8,150 tokens) · this version ~2,080 tokens · JSON https://www.anchorterminal.com/tools/gumloop.json · canonical https://www.anchorterminal.com/tools/gumloop
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-09

**C · 61.6/100 · rank #419 of 842 · #4 in Workflow automation · not agent-ready · confidence medium**


Assessment: Gumloop's REST API has a public OpenAPI file with 105 operations, Markdown docs and a changelog with 28 dated releases since 9 July 2026. API keys need the $37 Pro plan, the trial needs a card, the key can travel in the URL, and self-serve content may be used to train Gumloop's models.

## Facts

- Kind: HTTP API · vendor: AgentHub Inc. · category: Workflow automation · legal entity: AgentHub Inc. (doing business as Gumloop) · provenance 72/100
- Endpoint: `https://api.gumloop.com/api/v1` (HTTP, Streamable HTTP)
- Auth: OAuth or key · pricing: Paid · x402: no · licence: Proprietary service under Gumloop's terms of service. The Python SDK and CLI repository and the npm package are Apache-2.0
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The public REST API at https://api.gumloop.com/api/v1. The hosted MCP server, the CLI and the Python SDK call the same service and are described here but not graded apart
- API: OpenAPI 3.0.0 `Public API` 1.0.0 at https://docs.gumloop.com/api-reference/openapi.yaml, 105 operations on 83 paths. Agents, sessions, triggers, skills, artifacts, MCP servers, Company Brain, evaluations, models, chat completions, organisation administration and seven legacy flow endpoints such as `/start_pipeline`
- Auth: Personal or team API key as a bearer token (a personal key also needs `x-auth-key` with the user ID), or an OAuth 2.0 access token from the authorisation code grant with PKCE. Scopes are `gumloop_api` and `userinfo`. Keys and the `gumloop_api` scope need Pro or above
- MCP server: https://mcp.gumloop.com/gumloop/mcp, Streamable HTTP with OAuth sign-in, 46 tools for agents, sessions, skills, artifacts, connected apps, Brain search, flows, organisation and docs search
- Webhook triggers: `POST https://api.gumloop.com/trigger_incoming_webhook/<trigger_id>/<secret>` with no other credential. 200 with `{"success": true}` on acceptance, 404 for an unknown or disabled trigger or a wrong secret, 100 requests a minute per trigger with `Retry-After` on 429
- Sessions: `POST /agents/{agent_id}/sessions` returns 202 with the session `processing` or `queued`, or 201 for an idle session. States include `approval_required`. Streaming is Server-Sent Events on `ws.gumloop.com`
- Pagination: `page_size` and `cursor` with `next_cursor` on 13 list operations, plus `search`, `team_id`, `status` and `sort_order` filters
- Errors: 401 documented on 103 operations, 403 on 99, 404 on 77, 400 on 63, 409 on 16 and 429 on 5. No shared error schema in the OpenAPI file. MCP tool calls return a per-result `status` and an `error` with `code`, `message` and `type`
- Rate limits: Concurrency, not request rate. 25 concurrent agent interactions on Pro and 100 on Enterprise, across the organisation. At the limit Pro gets HTTP 429 with `gummie_rate_limit` and Enterprise requests are queued with a `queue_position`
- Approvals: Per connector, Always allow (the default), Ask each time, Ask for writes/deletes, Never allow or Custom per tool, plus App Rules written as CEL conditions. `POST /sessions/{session_id}/approvals` resolves a pending ask
- SDKs: Python `gumloop` 0.5.5 (8 October 2026, Apache-2.0, Python 3.10 or later). JavaScript `gumloop` 1.0.2 (30 December 2024), flows only. CLI installed by a shell script from gumloop.com
- Billing: Credits at $0.005 each. A run is model cost at list price, 1 credit per successful connector call plus any tool charge, 5 credits per session-minute of compute, and an 8 per cent orchestration fee (16 per cent with your own model key). Failed connector calls are not charged
- Releases: Numbered releases on https://www.gumloop.com/changelog, 12.1.0 on 7 October 2026. 3 in October, 9 in September, 9 in August and 11 in July 2026
- Compliance: The security page claims SOC 2 Type II attestation, HIPAA compliance with BAAs on eligible plans, and certification under the EU-U.S. Data Privacy Framework. DPAs are on request for Enterprise. The trust centre at trust.gumloop.com is drawn by script and was not read
- Prices: Pro plan $37 per month (plan); Credit (overage and list price) $0.005 per credit; Agent compute $1.50 per session-hour; Connector tool call $0.005 per call

- Scores: Reliability 82, Performance pending, Schema & documentation 83, Agent ergonomics 66, Security & auth 50, Payments & pricing 20, Task success pending, Maintenance & community 77, Transparency & trust 57 · negative events -2 · total over the 7 assessed categories
- Why: Reliability, Graded as a hosted service on the REST API. · Schema & documentation, Public OpenAPI 3.0.0 file with 105 operations on 83 paths (25). · Agent ergonomics, List operations take `page_size`, and a session read returns its whole message list. · Security & auth, Personal and team API keys, and OAuth 2.0 with PKCE, refresh tokens and a revoke endpoint, where the one API scope, `gumloop_api`, grants th… · Payments & pricing, No x402, MPP or L402 (0). · Maintenance & community, Release 12.1.0 is dated 7 October 2026 and the Python SDK 0.5.5 was published on 8 October (30). · Transparency & trust, Closed service under published terms with AgentHub Inc., and the SDK is Apache-2.0 (15).
- Sources: 27, open questions: 8, both in the full twin
- Capabilities: automation.workflows, automation.apps, automation.webhooks, agent.tools
- JSON: https://www.anchorterminal.com/api/v1/tools/gumloop.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/gumloop.svg` or a link to https://www.anchorterminal.com/tools/gumloop from a page on gumloop.com or one of its subdomains, or the README of github.com/gumloop/gumloop-py, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Send the key as `Authorization: Bearer` and, for a personal key, add the `x-auth-key` header with the user ID. Do not put `api_key` in the URL
2. Create sessions with `POST /agents/{agent_id}/sessions`, then poll `GET /sessions/{session_id}`. A 202 means processing or queued, and the reply is not in the response
3. Supply your own `session_id` when creating a session. A repeat returns 409, which makes a retry safe
4. For streamed output send the same body with `stream: true` to `ws.gumloop.com`. `api.gumloop.com` answers 400 to it
5. Treat a webhook trigger URL as a password. It returns `{"success": true}` at once and never carries the agent's output

## Connect

```bash
uv add gumloop
```

```bash
curl -X POST 'https://api.gumloop.com/api/v1/agents/abc123DEFghiJKL/sessions' -H 'Authorization: Bearer YOUR_ACCESS_TOKEN' -H 'Content-Type: application/json' -d '{"input": "Research Acme Corp and draft a brief."}'
```
Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/gumloop

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| Pipedream API + MCP | B | 65.5 | automation.workflows, automation.apps, automation.webhooks, agent.tools | https://www.anchorterminal.com/tools/pipedream.min.md |
| Kestra | B | 63.6 | automation.workflows, automation.webhooks, automation.apps, agent.tools | https://www.anchorterminal.com/tools/kestra.min.md |
| Make API + MCP | C | 58.7 | automation.workflows, automation.apps, automation.webhooks, agent.tools | https://www.anchorterminal.com/tools/make.min.md |
| Workato API + MCP | C | 58 | automation.workflows, automation.apps, automation.webhooks, agent.tools | https://www.anchorterminal.com/tools/workato.min.md |
| Activepieces API + MCP | C | 57.5 | automation.workflows, automation.apps, automation.webhooks, agent.tools | https://www.anchorterminal.com/tools/activepieces.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)