# Fix list: Sendblue From Anchor Terminal's listing at https://www.anchorterminal.com/tools/sendblue, the October 2026 research run, assessed 8 October 2026. Grade C, 61.7 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 Sendblue: 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. Security & auth, 46 out of 100, up to 9.5 more on the total Why it scored 46: One key pair per account in two headers, never in the URL, plus temporary bearer tokens that can be limited to named phone numbers, expire after 900 seconds by default and can be revoked. Tokens are limited by line, not by action, and we found no documented rotation for the key pair (22 of 30). No read-only key. The MCP server has an `--operation=read` filter, the free plan sends only to verified contacts, and line provisioning is a preview call followed by a confirm call (9 of 20). Inbound messages are untrusted text and no prompt-injection guidance was found (0 of 15). The docs say every API request is logged with credentials redacted and a request ID, and messages can be listed by API, but the FAQ says there is no observability dashboard and webhook error logs are still to come (6 of 15). Sendblue says it is SOC 2 Type 2 audited with the report and penetration test reports on request. No security.txt, disclosure policy or bug bounty was found (9 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. ## 2. Payments & pricing, 40 out of 100, up to 7.5 more on the total Why it scored 40: No x402, MPP or L402 (0). Plan prices are public, $0 and $100 a month per line with no per-message fee, while the outbound plan is priced through sales (10 of 20). The free plan needs no card (20). The CLI creates an account and API keys with no browser, but a person has to text a one-time code from a phone or supply an email code, so half (10 of 20). 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. Reliability, 71 out of 100, up to 5.8 more on the total Why it scored 71: Read with the hosted lines. status.sendblue.com is an UptimeRobot page with five monitors and 90 days of daily ratios (20). No incident is written up there. The API ingestion health monitor shows 95.349 per cent for 16 July 2026 (about 67 minutes, and we can't tell whether it was one outage), 98.770 per cent for 24 July, and API root 98.192 per cent for 7 October, so between minor incidents and one major (15 of 30). Rate limits are published with numbers per plan and per line (15). The docs say to slow down or spread traffic on a 429 and to verify state after `504 outcome_unknown`, with no Retry-After header documented and an idempotency key only on App Card updates (8 of 15). The API page shows a 99.9% Uptime SLA figure and the docs list custom SLAs for enterprise, but no SLA document was found and the terms disclaim availability (3 of 10). The REST API is generally available (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. ## 4. Schema & documentation, 66 out of 100, up to 5.5 more on the total Why it scored 66: We graded the REST API. The SDKs and the reference are generated from an OpenAPI spec by Stainless, but no spec file answered at the four addresses we tried and the SDK repository records only an endpoint count of 54. The published MCP package has JSON Schema inputs on its 18 tools (12 of 25). `llms.txt` and a Markdown twin of every docs page at `/index.md` (10). The reference and `llms.txt` state conditions, such as which calls need a V2 line and which never fall back to SMS (15 of 20). Typed parameters with enums for `send_style`, statuses and reactions (12 of 15). Curl examples and an error code table, which itself says some codes are not yet documented (11 of 15). Paths mix `/api`, `/api/v2`, `/v3` and `/accounts`, there is no API changelog in the docs, and the SDK changelog mostly records regenerated clients (6 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. ## 5. Agent ergonomics, 76 out of 100, up to 3.9 more on the total Why it scored 76: Message lists take `limit` from 1 to 100, and the MCP server adds a `jq_filter` on every tool, a read-only filter and a three-tool dynamic mode (20 of 25). `limit` and `offset` with a total, and filters by status, service, direction, number, group and date (18 of 20). Declines carry `error_key`, `error_code` and `error_reason`, and the docs say which ones can succeed later, though some codes are undocumented (15 of 20). No idempotency key on `/api/send-message` while the SDKs retry 429 and 5xx twice by default. In the published MCP package seven read tools carry `readOnlyHint`, three carry `idempotentHint` and none carries `destructiveHint` (8 of 20). A send needs three fields, and there are official TypeScript and Python SDKs (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. ## 6. Transparency & trust, 58 out of 100, up to 3.7 more on the total Made of editorial 29, provenance 86. Why it scored 58: Closed service with clear terms from Round One, Inc., and Apache-2.0 SDKs (15 of 30). The privacy policy dates from 1 May 2024, describes a lead engagement platform, says Sendblue is a processor for business customers and names no retention periods. The docs say inbound media links expire after 30 days. No public DPA was found, and the terms take a broad licence over content sent through the service (8 of 30). No deprecation policy. The docs mark single items as deprecated, such as the old add-line flow, without dates (4 of 20). No sub-processor list or data locations. The docs name Twilio and Agora for calling only (2 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): - Terms of service: read, states 5 of the 7 things a reader expects (8.3 of 10) - Privacy policy: read, states 5 of the 8 things a reader expects (7.8 of 10) - security.txt: not found (0 of 10) ## 7. Maintenance & community, 72 out of 100, up to 2.5 more on the total Why it scored 72: `sendblue` 3.19.1 on npm and 1.31.1 on PyPI on 7 October 2026 (30). Eight npm releases between 22 September and 7 October alone (20). No API changelog in the docs, support by email, and we couldn't read the GitHub issue trackers (6 of 15). Official TypeScript and Python SDKs are current, but `sendblue-api-mcp`, the MCP package the docs install, is still 2.0.1 from 5 August 2025 and no Sendblue messaging server is in the official MCP registry (10 of 15). The SDK repository has CI and release workflows, and the published MCP package pins the SDK at `^2.0.1` (6 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. ## 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 stars and open issues for sendblue-api/sendblue-ts, because the GitHub API refused us for its rate limit - unchecked: whether CI passes on the SDK's default branch, which a clone doesn't show - unchecked: the SOC 2 Type 2 report and penetration test reports, which Sendblue supplies on request - No public OpenAPI file was found at `/openapi.json`, `/openapi.yaml`, `/openapi.yml` on docs.sendblue.com or `/openapi.json` on api.sendblue.com - Whether the 67 minutes missing from the API ingestion monitor on 16 July 2026 were one outage. The status page has no written incident for it - The pricing page says the free plan has no outbound messaging and no webhooks, while the docs say free accounts can message verified contacts. We didn't test which holds - Whether the 99.9% Uptime SLA figure on the API page is backed by a contract term. No SLA document was found - Which countries' numbers Sendblue can assign and reach. The docs describe E.164 numbers and area codes without a coverage list - Whether the account key pair can be rotated or more than one pair issued - The lead's interface line reads REST API (v2). The API has no single version, with paths under `/api`, `/api/v2`, `/v3` and `/accounts` ## Weaknesses - Proactive outbound messaging needs the Blue Ocean plan, which has no public price. The $100 AI Agent plan is inbound-first - No OpenAPI file was found at any public address, although the SDKs and the reference are generated from one - The docs install `sendblue-api-mcp`, last published on 5 August 2025 with 18 tools. The API now has 54 configured endpoints - `/api/send-message` takes no idempotency key, and the SDKs retry 429 and 5xx responses twice by default - No public DPA, sub-processor list, retention period, deprecation policy or security.txt was found ## 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. - Send `from_number` on every request. Read the account's numbers from `GET /api/lines` first, because a send without it fails - On the free plan a recipient must text the Sendblue number once before any message to them is accepted - Before a contact's first reply keep messages under 300 characters with no links, media, phone numbers or email addresses, and stop at six - Branch on `error_key` in an HTTP 400 decline and don't retry it. Only HTTP 429 clears by waiting - Set `maxRetries: 0` on sends where a duplicate matters, and treat `504 outcome_unknown` on group changes as possibly applied - Mint a token at `/v3/auth/tokens` limited to one phone number for a sub-agent, and keep the account key pair out of its context ## 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.