Zapier MCP (agent actions) by Zapier

# Fix list: Zapier MCP (agent actions)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/zapier-mcp, the October 2026 research run, assessed 1 October 2026. Grade C, 58.4 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Zapier MCP (agent actions): work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Reliability, 48 out of 100, up to 10.4 more on the total

Why it scored 48: Statuspage at status.zapier.com with an MCP component and an incident feed (20). No incident in the last 90 days names MCP, but Webhook Triggers and Zap Actions, including Searches & Writes, were disrupted for about 1 hour 35 minutes on 8 September 2026, 401s hit several products for about 1 hour 15 minutes on 13 July, and the feed lists 20 more app-specific incidents (10). No numeric rate limits, the docs say there's no per-session limit and the task allowance is the only cap (5). At the cap calls stop with an error, with no Retry-After or retry guidance (3). No SLA found (0). The MCP server is GA (registry 1.0.1 on 29 July 2026), the SDK is in open beta (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 2. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). Each successful call costs two tasks and plan prices are public, so the per-call cost can be worked out but depends on the plan (15). Free plan with 100 tasks a month and no card, and MCP is available to every account type (20). A person signs up in a browser and connects apps. MCP Embed lets a product create servers for its users, which still needs a human sign-in (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 3. Security & auth, 56 out of 100, up to 7.7 more on the total

Why it scored 56: OAuth for listed clients, otherwise a long-lived connection token tied to one server, revoked by regenerating it. The docs document `?token=` in the URL as a working option while preferring the header, so we deduct 10 (12). Account-level app and action restrictions apply, managed mode limits the agent to chosen actions, admins can switch MCP off per workspace and set task quotas, but there's no approval step for writes (15). Actions return email, CRM and document content with no prompt-injection guidance (3). User-level activity logs for every tool call in the History tab, which are deleted with the server (13). SOC 2 Type II and SOC 3 and a bug bounty on the security page, no platform named, no ISO 27001, and no security.txt per the 30 September check (13).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 4. Schema & documentation, 64 out of 100, up to 5.9 more on the total

Why it scored 64: MCP tools carry JSON Schema by protocol, but the 16 meta-tools are server-side and action schemas are generated per app, so we couldn't read them (15). llms.txt with .md pages for every MCP doc (10). The docs say when to use the read and write execution tools, while action schemas are terse (12). Action input typing not visible to us (7). Troubleshooting guides cover MCP errors, with few worked examples of calls (10). Registry versions and an SDK changelog with stability tiers, but no dated changelog for the hosted MCP server itself (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 5. Agent ergonomics, 67 out of 100, up to 5.4 more on the total

Why it scored 67: 16 fixed meta-tools in agentic mode, which discover, enable and run actions at runtime, or managed mode with only pre-selected actions (25). The SDK pages list results with cursors, but we found no output-size controls on MCP action results (12). Errors come from the app behind each action, and troubleshooting guides cover the common MCP failures (12). Reads and writes run through separate tools, but we found no annotations or idempotency guidance, and a batch of five rows is five separate calls (10). The SDK is TypeScript only and in beta (8).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 6. Maintenance & community, 76 out of 100, up to 2.1 more on the total

Why it scored 76: SDK release on 29 September 2026 per the 30 September check, at 0.113.0 for the TypeScript SDK and 0.85.0 for the CLI (30). The changelog lists versions 0.79.0 to 0.113.0 without dates, so we can't count releases in the last 90 days ourselves and gave partial credit (15). Closed service with an SDK changelog, support and a send_feedback tool in the server, and no public issue tracker (10). Registered as com.zapier/mcp 1.0.1 in the official MCP registry (15). The SDK packages are current on npm but still in beta (6).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## 7. Transparency & trust, 76 out of 100, up to 2.1 more on the total

Made of editorial 61, provenance 90.

Why it scored 76: Closed service under Zapier's terms (15). Data stored in AWS us-east-1, configurable retention only on Enterprise, and only Enterprise is opted out of AI training by default, which leaves other plans' position unstated on the MCP security page (18). AI Actions and NLA retired with a pointer to MCP, the SDK attaches deprecation notices to tool results and has stated stability tiers (16). Data location stated, subprocessor list not checked in this run (12).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- security.txt: not found (0 of 10)

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- Whether non-Enterprise MCP data can be used for AI training and how to opt out.
- Dates of SDK releases, which the changelog page we read doesn't show, so the last release relies on the 30 September check.
- unchecked: the subprocessor list and the bug bounty's platform.
- Whether the 8 September 2026 Searches & Writes disruption affected MCP calls.

## Weaknesses

- Connection tokens are long-lived and the docs allow them in the URL query string
- No numeric rate limits and no retry guidance when the task cap is hit
- Only Enterprise is opted out of AI training and gets configurable retention
- Batch work counts per item, so five rows written is ten tasks
- Streamable HTTP only, SSE-only clients can't connect

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Call discover_zapier_actions and enable_zapier_action before executing, discovery is free and execution costs two tasks
- Use execute_zapier_read_action for lookups, keep execute_zapier_write_action for changes
- Send the connection token in the `Authorization` header, never in the URL
- Count batch writes per item before starting, five rows is ten tasks

## What the review panel asked for

- A programmatic sign-up route
- short-lived tokens
- drop the query-string token

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.