Respan API + MCP by Respan (formerly Keywords AI)

# Fix list: Respan API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/respan, 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 Respan 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402 or other machine payment (0). Overage prices published without login, $8 per 100,000 logs and $1 per 1,000 scores on Team (20). The Free plan has 100,000 logs and unlimited seats. The pricing page doesn't say whether a card is needed, so half (10). Keys can be minted by API, but only with an existing Admin key, so a person still signs up in a browser first (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.

## 2. Security & auth, 66 out of 100, up to 6 more on the total

Why it scored 66: API keys carry Read, Write, Gateway and Admin scopes since 25 September, can have an expiry and spend caps, are bound to one project and stop working at once when revoked. The MCP server also takes OAuth (30). A Read-only key plus the tool allowlist gives a read-only agent, but the write and delete tools have no confirmation step of their own (15). Logs hold whatever the application sent, and we found no prompt-injection guidance for agents reading them (3). Every gateway call is logged against its key. The compliance page mentions audit logging without saying what customers can see (8). SOC 2 Type II (report under NDA), HIPAA BAAs and GDPR per the compliance page, and a trust centre at trustcenter.respan.ai that we didn't load. No security.txt or SECURITY.md, and security fixes appear in the changelog as one-liners (10).

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, 66 out of 100, up to 5.5 more on the total

Why it scored 66: The platform MCP server registers 67 tools in its source, about 24,000 characters of descriptions before schemas (5). The `Respan-Enabled-Tools` header trims the list server-side (8 back). `list_logs` pages 1 to 50 rows (default 20), returns summary fields unless `include_fields` asks for more, and defaults to the last hour (20). API errors pass through with status and body, and invalid filters return a typed `validation_error` naming the field (15). No `readOnlyHint` or `destructiveHint` on any tool, although `delete_dataset`, `delete_evaluator`, `delete_experiment`, `delete_workflow` and `remove_dataset_logs` are among them, and no idempotency keys found (3). Python and TypeScript SDKs, a CLI and agent skills, with few required parameters (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. Reliability, 73 out of 100, up to 5.4 more on the total

Why it scored 73: Better Stack page at status.respan.ai with six components (web, logging and gateway, each for the platform and Enterprise) and an incident history page (20). The last 90 days show two incidents, both in late September. The Responses API path through the gateway was affected for OpenAI and Anthropic requests for 13 hours 47 minutes on 26 September, and request logs were delayed across all regions for 1 hour 58 minutes on 29 September. The uptime bars still read 100 per cent for every component (10, one major). Gateway and ingestion limits published per plan, 412 a minute on Free and 8,400 and 4,200 on Team (15). The OpenAPI spec declares 429 on several operations. We found no `Retry-After` or backoff guidance (10). The pricing page lists a 99.99 per cent uptime SLA on Enterprise, with terms not published (8). Gateway, logging and MCP are GA (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.

## 5. Transparency & trust, 61 out of 100, up to 3.4 more on the total

Made of editorial 47, provenance 75.

Why it scored 61: Closed platform. The SDK monorepo is Apache-2.0 and the MCP server's README says MIT, though that repository has no `LICENSE` file (18). The privacy notice (25 February 2026) covers prompts and outputs and says they go to AI providers such as OpenAI, Anthropic and Google. Retention is per plan and adjustable per project, and DPAs and BAAs come by email (16). No deprecation policy found. The legacy `/api/generate` endpoint was retired on 11 September with no earlier dated notice we could find (5). EU residency on request and AI providers named, no full subprocessor list (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):

- Domain age: respan.ai, registered 2026-01-13 (under a year) (0 of 15)
- security.txt: not found (0 of 10)

## 6. Schema & documentation, 89 out of 100, up to 1.8 more on the total

Why it scored 89: OpenAPI 3.1 at respan.ai/docs/openapi.json with 40 or more paths, and every MCP tool takes a zod schema (25). llms.txt, per-section llms.txt and `.md` for every page (10). MCP descriptions say when to use a tool and when not to. `list_logs` tells the model to filter server-side rather than fetch everything and to call `get_log_detail` for full data, and `delete_dataset` says it can't be undone (17). Typed inputs with operator enums, but `page_size` bounds sit in the description rather than the schema and filter values are `any` (11). Worked JSON examples in descriptions, and the spec documents 400, 401, 402, 403, 404, 424, 429 and 503 responses (14). Weekly dated changelog. API paths mix `/api/` and `/api/v1/` with no version policy (12).

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.

## 7. Maintenance & community, 85 out of 100, up to 1.3 more on the total

Why it scored 85: Packages published from the Apache-2.0 monorepo on 2026-09-30 (30). 195 commits and well over three package publishes since 3 July, and 13 dated changelog entries since 7 July (20). The MCP repository last changed on 15 September. We didn't load issue reply times. Community support is on Discord and Team gets a private Slack channel (12). Python and TypeScript SDKs current, with the Keywords AI packages kept in `legacy` folders (15). CI and an OpenAI Python compatibility workflow run in public (8).

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.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 2026-07-30 and 2026-08-28. The changelog lists "Closed an unauthorized live-event stream vulnerability" and "Patched some security vulnerabilities" with no advisory, scope or statement on whether customer data was exposed. Fixed, but disclosed in one line each. -2 (https://www.respan.ai/docs/changelog/changelog)

## 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 the 30 July "live-event stream" fix exposed other customers' data, which the changelog doesn't say
- Whether the legacy `/api/generate` endpoint had a deprecation notice before its retirement on 11 September
- Whether a card is needed for the Free plan
- unchecked: the trust centre at trustcenter.respan.ai, so the subprocessor list and SOC 2 report date are unverified
- Why the status bars show 100 per cent for every component despite the 26 and 29 September incidents

## Weaknesses

- 67 MCP tools load by default, with delete tools and no `readOnlyHint` or `destructiveHint` annotations
- A 13-hour 47-minute gateway incident on 26 September 2026, while the status bars show 100 per cent
- Security fixes on 30 July and 28 August 2026 disclosed only as one-line changelog entries
- Free plan keeps logs for 7 days
- No deprecation policy, and the legacy `/api/generate` endpoint was retired on 11 September 2026

## 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.

- Create a key with only the Read scope for agents that analyse logs. New keys get Read, Write and Gateway by default
- Set `Respan-Enabled-Tools` to a short list such as list_traces,get_trace_tree,list_logs. All 67 tools load otherwise
- Pass `filters` to `list_logs` rather than paging through everything. The default window is the last hour
- Call `list_organizations` first when results look empty over OAuth. Every tool acts on the active organisation only
- Append `.md` to any docs URL for clean Markdown

## What the review panel asked for

- a written deprecation policy
- annotate delete tools
- bounds in the schema

## 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.