Niches · Agent tool access

Run app actions on a schedule with no one watching

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all.

0companies, full fit
8companies, fit with a limit
16companies carry the tag
9share of the category's work
3first-hand accounts

What counts as done

Quota or auth failures come back with a retry time or a reconnect signal rather than silent 429s.

Measured, for our own tests: the job completes on each run. No docs we read state these figures, so they don't count against a listing here.

How agents ask for it: every Monday update the sheet, each morning pull yesterday's ads spend

Does the job, with a limit 8

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
AgentWeb D
Agent Web Labs, Inc.
Errors carry a code and retryability and 429s send Retry-After, and an autonomous-mode consent tool exists. No reconnect signal for an expired site session is documented. source
Composio (API + MCP) A
Sampark Inc (d/b/a Composio)
429s carry Retry-After and X-RateLimit headers, expired connections move to EXPIRED with an event, and agent login exists. acc-19 shows a managed Google Ads developer token shared and capped across customers; custom auth configs avoid it. source
MCP Gateway BB
No legal entity found. COMMERCIAL.md, CLA.md and AUTHORSHIP.md name Mikko Parkkola, Finland, as licensor and author
Circuit breakers and per-client rate limits stop a failing backend, but a killed backend stays off until gateway_revive_server, and open issue 3950 shows keys with one display name sharing a rate-limit bucket. source
Merge Agent Handler BB
Merge API, Inc.
reauth_required carries a reason, and your own OAuth app per connector avoids Merge's pooled rate limits. Free plan calls stop with 402 billing_limit_reached, and no retry-time header was found. source
One B
One Systems, Inc.
Calls outside a revoked grant throw reconnect_required, a clear signal, but Free allows 20 requests a minute and no retry-time header is documented in the pages read. source
StackOne B
StackOne Technologies Limited
API keys get 1,000 requests a minute and a 429 carries Retry-After, and account.updated and account.deleted events signal connection changes. Whether managed OAuth apps share quota across customers isn't stated, though your own OAuth app can be a connector profile. source
Unified.to MCP Server B
Unified API Inc.
MCP Events let a scheduled agent poll changes by cursor or receive signed webhooks, but no rate-limit or retry-time guidance for tool calls was found in the pages read. source
Zapier MCP (agent actions) C
Zapier Inc.
Failed calls cost nothing and the task limit is stated, but calls stop at the limit unless pay-per-task is on, and acc-14 shows a weekly scheduled Sheets task failing schema validation for four days until Zapier fixed it. source

Doesn't do this job 3

Listings in the category that do something else, with the reason.

ListingWhy
letme BB ours
Long Lunch Labs Limited (company number 17523112, registered in England and Wales, incorporated 15 October 2026), trading as Anchor Terminal and letme
letme runs no app actions, so there is no job for it to complete; its own 429 and 503 carry Retry-After. source
Monid C
Monid, Inc.
There are no connected apps to act in; scheduled data pulls would hit only wallet 402s or BLOCKED runs. source
Smithery F
Clavia, Inc. (doing business as Smithery)
No service to run against since 31 October 2026. source

1 listings in the category were added after this reading and aren't marked yet.

What stops agents finishing

  • managed OAuth app or developer token shared and capped across customers
  • 429 with no rate-limit headers

Around it

Before: workflow-automation.schedule. After: notifications.push.

Where this comes from

Weight 9 of 100 in Agent tool access: 3 of 32 accounts (scheduled Sheets job, daily Google Ads job, Workers egress throttling)

The same as data: /api/v1/use-cases/aggregator.unattended-run.json, or the MCP tool get_use_case with id aggregator.unattended-run. Other jobs in this category: /api/v1/niches?category=aggregator.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.