Salsa

# Fix list: Salsa

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/salsa, the October 2026 research run, assessed 8 October 2026. Grade D, 46.1 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 Salsa: 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, 36 out of 100, up to 12.8 more on the total

Why it scored 36: Read with the hosted rubric, for embedded-payroll partner access. No status page is linked from salsa.dev or docs.salsa.dev, and status.salsa.dev didn't connect. salsa.statuspage.io exists with two components marked "(example)" and a last update of 4 November 2022, which we didn't count (0). No readable incident history (5). Limits are published as 3,000 requests every 5 minutes per IP address and per API token (15). The docs name the 429 response without a Retry-After header or backoff guidance. There are no idempotency keys, though a create that repeats an externalId fails with a uniqueness error (6 of 15). No SLA found (0). REST v1 with a production host, not marked 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, 5 out of 100, up to 11.9 more on the total

Why it scored 5: Read with the hosted rubric. No x402, MPP or L402 (0). No public prices, and www.salsa.dev/pricing returns 404 (0). No free tier or trial found. A sandbox exists, issued by Salsa after a contact form, so partial credit (5 of 20). A person contacts sales and Salsa issues the tokens (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. Agent ergonomics, 48 out of 100, up to 8.5 more on the total

Why it scored 48: No field selection on REST responses, and page size defaults to 500 (10 of 25). page, size and sort exist on four list endpoints, payroll runs take six filters, and the filter syntax is documented. Other list endpoints take no paging parameters (13 of 20). Errors carry a message, a target and a details array without machine-readable codes (10 of 20). No Idempotency-Key. Creates can carry an externalId that rejects duplicates, and a payroll run is previewed, created as PENDING, confirmed in a separate call and deletable while pending (10 of 20). No server SDK. The docs point to OpenAPI Generator, and the mock-onboard endpoints shorten sandbox setup (5 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.

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

Why it scored 56: One partner API token per environment that the docs say grants many privileges, plus user tokens with 12 roles in three tiers, a named employer list of up to 125 and a 60-minute default lifetime. No rotation procedure for the partner token was found (20 of 30). Basic roles can't add bank accounts, unmasking needs super-admin, limits are enforced at the API and payment needs a separate confirm call. No read-only role was found (12 of 20). Responses carry text entered by employers and workers, with no injection guidance (3 of 15). Audit entries name the actor, the entity, the action and the time (15). A SOC 2 Type 2 badge image on the home page. No security page, security.txt, disclosure policy or bug bounty found (6 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.

## 5. Schema & documentation, 73 out of 100, up to 4.4 more on the total

Why it scored 73: Public OpenAPI 3.1 spec with 123 operations and 78 webhook events (25). llms.txt with 292 lines, and every docs page has a Markdown copy (10). 111 of 123 operations carry a description and the spec holds about 3,200 description fields. They state purpose and seldom say when not to use an endpoint (13 of 20). Inputs use enums, required fields, formats and discriminated unions, with no free-form JSON bodies found (12 of 15). 364 field examples and worked curl tutorials. Every operation documents 400, 404, 422 and 500 with one envelope of message, target and details, with no error codes, and 401, 403 and 429 are missing from the spec (8 of 15). The path is versioned as v1. No API changelog was found, docs.salsa.dev/changelog returns 404, and the product updates blog stops at 11 February 2026 (5 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. Maintenance & community, 50 out of 100, up to 4.4 more on the total

Why it scored 50: The OpenAPI spec was last modified on 7 October 2026 and a docs page was updated the same day (30). There's no dated API changelog. 11 guide pages carry update dates between 27 July and 7 October 2026, which we counted as half (10 of 20). Closed service with support through a partner's Salsa contact, no public issue tracker or community, and a product updates blog that stops at 11 February 2026 (3 of 15). The only official package is the browser library @salsa-payroll/salsa-js 0.3.0 from 11 December 2025, with no server SDK (4 of 15). That package is published from GitHub Actions and its repository isn't public (3 of 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.

## 7. Transparency & trust, 51 out of 100, up to 4.3 more on the total

Made of editorial 31, provenance 70.

Why it scored 51: Closed service. The public Terms of Use are dated 7 April 2022 and cover the website and services in general, and the partner agreement for API use isn't public (12 of 30). The privacy policy of 24 September 2025 excludes data processed for platforms, gives no retention periods and still has a vendor name in square brackets. No DPA was found (10 of 30). The spec marks four operations deprecated and the docs list three deprecated roles with replacements, with no removal dates or policy (6 of 20). The policy says Salsa is headquartered in the United States. No subprocessor list or data locations were found (3 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):

- Domain age: salsa.dev, registered 2021-08-06 (5 years) (11 of 15)
- Terms of service: read, states 6 of the 7 things a reader expects, and has 2 clauses that cost points (5.1 of 10)
- Privacy policy: read, states 6 of the 8 things a reader expects (8.5 of 10)
- Status page: not found (0 of 10)
- 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.

- unchecked: whether status.salsa.dev exists. The host didn't connect from our network, and no status page is linked from the site or docs
- Who owns salsa.statuspage.io. It shows example components and a last update of 4 November 2022
- Prices, contract terms and any SLA, which sit in the partner agreement
- The GraphQL API's schema and whether it is open to every partner. The public index has no GraphQL reference
- Whether a partner can rotate or revoke its API token without contacting Salsa
- The scope and date of the SOC 2 Type 2 report shown as a badge
- Whether 429 responses carry a Retry-After header. We had no token to test with

## Weaknesses

- API tokens come from Salsa after a sales contact. No self-serve signup, public price, free tier or trial was found
- No status page, SLA, API changelog or deprecation policy was found on salsa.dev or docs.salsa.dev
- No Idempotency-Key header. The rate-limit page names 429 with no Retry-After or backoff guidance
- No server-side SDK. The only official package is the browser library Salsa.js, last published on 11 December 2025
- The signup link used across the docs, www.salsa.dev/get-a-demo, returned 404 on 8 October 2026
- No security.txt, disclosure policy, DPA or subprocessor list was found. SOC 2 Type 2 appears only as a badge image

## 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.

- Use the sandbox token against https://api.sandbox.salsa.dev and the production token against https://api.salsa.dev. Each environment has its own token
- Call POST /payroll-runs/preview before creating a run, then confirm the PENDING run in a separate call. Confirming starts the employer debit
- Send your own externalId on every create. A repeat returns a uniqueness error, which is the only duplicate protection
- Make sure externalId values are unique across all employers, since each entity type has one namespace
- Mint a user token with the lowest role and the fewest employerIds the task needs, and keep the partner token on the server

## 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.