Twenty API + MCP by Twenty

# Fix list: Twenty API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/twenty, the October 2026 research run, assessed 1 October 2026. Grade B, 65.9 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 Twenty API + MCP: 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, 46 out of 100, up to 10.8 more on the total

Why it scored 46: No status page found. status.twenty.com returns the Twenty app shell, and the website's links name only a trust centre (0). No readable incident history (5). Cloud limits published, 100 requests a minute and 60 records per batch (15). 429 handling isn't documented on the API page. GraphQL batch upserts and the MCP `upsert_many_*` tools give a safe retry for creates (6 of 15). The Enterprise plan lists 'Dedicated support & SLA' in the website source (10). REST, GraphQL and MCP carry no beta label (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, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Plan prices public, Cloud Pro $9 or $12 and Organization $19 or $25 a user a month, but nothing per call (10). Self-hosting is free, and the server's billing defaults allow a 7-day cloud trial without a card (20). A person signs up in a browser or deploys a server (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, 58 out of 100, up to 7.4 more on the total

Why it scored 58: API keys are Bearer tokens with a required expiry date, a revoked flag and an optional role binding. OAuth with dynamic client registration for MCP and apps. Tokens stay out of URLs (30). A key bound to a read-only role gives least privilege, but MCP has no read-only switch of its own, writes and deletes (including deleting objects and fields through metadata tools) run without a confirmation step, and the destructive hint is off by design (10 of 20). Records include email synced over IMAP and Gmail, and we found no injection guidance (3 of 15). The subprocessor list says audit logs are kept in ClickHouse. We didn't find which plans see them (7 of 15). security.txt with a contact and a policy link, but no Expires field. No SOC 2 or bug bounty found, the trust centre renders client-side, and an open bug (#26212, 18 September 2026) reports the /dpa redirect showing the workspace sidebar to signed-out users (8 of 20).

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. Agent ergonomics, 80 out of 100, up to 3.3 more on the total

Why it scored 80: By default the MCP server lists six meta-tools (`execute_tool`, `learn_tools`, `load_skills`, `list_object_metadata_names`, `list_skills`, `get_tool_catalog`) and loads schemas on demand. `?mode=direct` lists every tool for clients that load lazily (25). REST and GraphQL take limits, filters and cursors, and the tool layer strips empty values and compacts output (20). `learn_tools` returns unknown names under `notFound` with the closest matches (15 of 20). Read-only tools carry `readOnlyHint`. `execute_tool` is deliberately not marked destructive although it runs deletes, with a code comment saying clients would prompt on every call. `upsert_many_*` exists for safe retries (10 of 20). Few required parameters. TypeScript SDKs only (`twenty-sdk`, `twenty-client-sdk`) (10 of 15).

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.

## 5. Schema & documentation, 88 out of 100, up to 2 more on the total

Why it scored 88: Each workspace serves its own OpenAPI files at `/rest/open-api/core` and `/rest/open-api/metadata`, generated from its schema, custom objects included (25). llms.txt and Markdown docs (10). The MCP server sends instructions that explain the tool-name grammar, when to use `get_tool_catalog` and that workflow and metadata tools need their skill loaded first (18 of 20). Inputs are typed from the workspace schema and `learn_tools` returns them on demand (13 of 15). Docs examples and JSON-RPC error codes. We didn't find a REST error reference (10 of 15). Release notes at twenty.com/releases and a breaking-changes CI check, but the API itself isn't versioned beyond the generated schema (12 of 15).

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.

## 6. Transparency & trust, 80 out of 100, up to 1.8 more on the total

Made of editorial 69, provenance 90.

Why it scored 80: AGPL-3.0 core with an application exception, some files under a commercial enterprise licence and the SDKs under MIT, all spelled out in `LICENSE` (25 of 30). Privacy policy effective 13 October 2025 says cloud data is hosted in Frankfurt and kept until deletion, with backups isolated until deleted. No retention periods in days (22 of 30). Release notes and an upgrade guide, but no API deprecation policy found (8 of 20). Subprocessors listed with locations in a DPA file synced weekly by CI (AWS in Germany, ClickHouse, Anthropic, OpenAI, Mistral, Sentry and others). Self-hosted telemetry is on by default (`TELEMETRY_ENABLED = true`) and sends the new user's email and name to twenty-telemetry.com on sign-up. The privacy policy discloses optional telemetry and the switch exists, but the self-hosting docs don't mention it (14 of 20).

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):

- Status page: not found (0 of 10)

## 7. Maintenance & community, 97 out of 100, up to 0.3 more on the total

Why it scored 97: twenty/v2.44.0 tagged on 30 September 2026 (30). Fourteen tags between 9 and 30 September 2026 (20). 60 open issues, and recent bug reports carry priority labels within days (22 of 25). Not in the official MCP registry (only a third-party io.usefulapi/twenty), but the official `twenty-sdk` is current at 2.43.0 (15). CI covers server, front end, SDK, end-to-end tests and breaking changes, with 50 commits on 1 October alone (10).

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.

## 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.

- unchecked: GitHub security advisories for twentyhq/twenty, the page was refused by our fetch rate limit
- unchecked: SOC 2 or other certifications, because trust.twenty.com renders client-side
- unchecked: pricing, privacy and status pages on twenty.com directly, refused by our fetch rate limit. Prices and the privacy policy come from the website source in the repository
- Whether the cloud trial runs with the server defaults (7 days without a card, 30 with one)
- The 30 September listing named status.twenty.com as the status page, but it serves the Twenty app. Patched to empty

## Weaknesses

- No status page found
- `execute_tool` isn't flagged destructive even though it can delete records, objects and fields
- Cloud limit of 100 requests a minute
- Self-hosted telemetry is on by default and sends sign-up emails and names
- No SOC 2 or bug bounty found, and the trust centre renders client-side

## 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 `learn_tools` with every tool name you need in one call before `execute_tool`
- Build CRUD names from the grammar (`find_many_companies`, `upsert_many_people`) instead of browsing the catalogue
- Pass one category to `get_tool_catalog`, since an unfiltered call returns hundreds of operations
- Load the matching skill before any workflow or metadata tool, as the server instructions require
- Batch up to 60 records a call to stay inside 100 requests a minute

## What the review panel asked for

- flag destructive operations
- document REST errors
- destructive hint on `execute_tool`
- confirmation for schema deletes

## 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.