Deel

# Fix list: Deel

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/deel, the October 2026 research run, assessed 7 October 2026. Grade B, 69.1 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 Deel: 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Per-person monthly prices are public, such as Deel HR from $5 per employee, with no API charge found. We scored it between plan-only and per-unit (15). The quickstart describes a sandbox created after signing up at app.deel.com, but the pricing page lists no free plan and we didn't confirm that signup needs no card or contract (10). A person signs up and creates tokens in a browser. Eight public MCP tools (hiring cost and salary data) are listed without a token, and OAuth clients can self-register (5).

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.

## 2. Security & auth, 64 out of 100, up to 6.3 more on the total

Why it scored 64: OAuth 2 with PKCE, dynamic client registration, single-use refresh tokens and about 90 scopes split into read and write. Organisation and personal tokens are scoped, revocable and have a sensitive-data setting (30). Read scopes allow a read-only agent and MCP tools carry annotations. No confirmation step for writes such as time-off rejection was found (13). The API returns free text written by workers and managers, and no prompt-injection guidance was found (3). An `api-logs:read` scope exists in the OAuth metadata, but no operator audit log for API or MCP calls was found in the docs (3). security.txt is valid until 21 July 2027, the CLI's security policy takes reports at security@deel.com, and deel.com/security lists SOC 1, SOC 2, SOC 3 and ISO 27001. No bug bounty found (15).

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.

## 3. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: The MCP docs list 86 tools with no toolsets. Unauthenticated clients see 8 public ones, and scopes limit what a token can call. REST lists take `limit`, and people has a `fields` filter (10). Pagination and filters exist on list endpoints but in three styles (offset, cursor, `page_size` with `next`) (16). Errors carry a message, the JSON path of the failed field, a request id and a docs link, and the MCP guide separates retryable from non-retryable errors (16). `Idempotency-Key` on POST and PATCH with a 24-hour cache, declared in the spec on one operation. The 8 MCP tools we could list all set readOnlyHint and destructiveHint (17). One TypeScript SDK, last published 13 March 2025, and a CLI at 0.2.0. No Python SDK found (6).

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.

## 4. Transparency & trust, 69 out of 100, up to 2.7 more on the total

Made of editorial 43, provenance 94.

Why it scored 69: Editorial half only. Closed service. The platform terms, privacy policy and DPA exist at stable URLs but returned no text without JavaScript, so we scored the terms as partly established. The CLI is MIT (10). Data handling couldn't be read for the same reason. The security page states AES-256 at rest and AWS hosting (5). The versioning policy gives at least one year stable and one year deprecated with `Sunset` headers, and the spec dates removals, such as the current people list on 30 September 2027 (20). Hosting locations are disclosed (Ireland, with disaster recovery in France). No sub-processor list was read (8).

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):

- Terms of service: published, but our reader couldn't read it (7 of 10)
- Privacy policy: published, but our reader couldn't read it (7 of 10)

## 5. Maintenance & community, 70 out of 100, up to 2.6 more on the total

Why it scored 70: The API changelog's newest entry is 5 October 2026 (30). 20 dated entries between 2 September and 5 October 2026 (20). Closed service with a dated changelog and a community forum at stack.deel.com, which we didn't read for response times (8). The TypeScript SDK was last published on 13 March 2025 and its repository isn't public. The CLI's first public release was 24 September 2026. The MCP server isn't in the official MCP registry (6). The CLI repository has a pull-request workflow and tests (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.

## 6. Reliability, 88 out of 100, up to 2.4 more on the total

Why it scored 88: Graded on the hosted REST API, with the MCP server on the same host. status.deel.com on StatusPal has components for the Deel Developer API and Demo & Sandboxes, with 60-day uptime bars (20). The incident history lists nothing after scheduled maintenance on 25 December 2025, and its last incidents are from July to September 2025 (30). The limit is published as 5 requests a second per organisation (15). Backoff guidance and an `Idempotency-Key` header for POST and PATCH are documented, but the rate limits page says no rate limit headers are returned while the MCP error page says 429 carries `Retry-After` (13). No SLA found. The platform terms didn't render without JavaScript (0). Version 2026-01-01 is labelled stable (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.

## 7. Schema & documentation, 85 out of 100, up to 2.4 more on the total

Why it scored 85: OpenAPI 3.1 specs are public under developer.deel.com/openapi/, 506 operations and 55 webhook events in the full file. The two spec links in the docs index return 404 (25). llms.txt, section indexes and a Markdown copy of each page (10). Operation descriptions average about 235 characters and name the required scopes, and some state limits such as rejection being irreversible. Few say when not to use an endpoint, and the public MCP tools have one-sentence descriptions (13). 1,988 enums, numeric bounds and formats in the spec, though several MCP tools wrap input in a single `data` object (13). 400, 401, 403, 404 and 500 are declared on nearly every operation with a shared error schema, 429 on only 28, and the spec carries few examples (9). Date-based versions and a dated changelog (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.

## 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: the privacy policy, data processing addendum and platform terms on deel.com/legal returned no document text without JavaScript, so retention periods, sub-processors, the contracting entity and any SLA weren't read
- unchecked: trust.deel.com is a JavaScript application and returned no content
- unchecked: whether signing up and creating a sandbox needs a card, a contract or a sales call. We didn't create an account
- unchecked: the authenticated MCP tool list. The server listed 8 public tools without a token, and the docs list 86 under different names
- unchecked: whether 429 responses carry `Retry-After`. The rate limits page says no rate limit headers are returned, the MCP error page says the header is sent
- unchecked: the OAuth access token lifetime, given as 1 hour on the MCP page and 30 days on the OAuth page
- unchecked: response times on the stack.deel.com community forum, and GitHub stars for letsdeel/deel-cli
- The legal entity Deel, Inc. is taken from the CLI's licence file, not from the terms
- No operator-facing audit log for API or MCP calls was found in the developer docs, though an api-logs:read scope exists
- The TypeScript SDK's repository (letsdeel/deel-api-js-sdk) isn't publicly cloneable

## Weaknesses

- 5 requests a second shared by every token in an organisation, with no rate limit headers on REST responses per the rate limits page
- The OpenAPI links in the docs index return 404. The working specs sit under developer.deel.com/openapi/
- The idempotency guide covers POST and PATCH, but the spec declares the `Idempotency-Key` header on one operation
- TypeScript SDK last published on 13 March 2025, and no Python SDK found
- Privacy policy, DPA and platform terms render only with JavaScript, and no public sub-processor list or SLA was read

## 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.

- Queue calls to stay under 5 requests a second for the whole organisation. Other integrations on the same account share the limit
- Send `X-Version: 2026-01-01` on REST calls, and watch `X-State`, `Deprecation` and `Sunset` response headers
- Send a UUID `Idempotency-Key` on POST and PATCH. Only retry a 5xx on a write when the request carried one
- Test against https://api-staging.letsdeel.com/rest with a sandbox token. Production and sandbox tokens aren't interchangeable
- Request read scopes only (people:read, time-off:read) for a read-only agent. Rejecting a time-off request is irreversible

## 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.