{
  "fixes": {
    "slug": "socure-riskos",
    "name": "Socure RiskOS",
    "listing": "https://www.anchorterminal.com/tools/socure-riskos",
    "markdown": "# Fix list: Socure RiskOS\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/socure-riskos, the October 2026 research run, assessed 8 October 2026. Grade B, 63.5 out of 100.\n\nThis 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.\n\nFor a coding agent working on Socure RiskOS: 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.\n\n## 1. Agent ergonomics, 46 out of 100, up to 8.8 more on the total\n\nWhy it scored 46: Scored on the REST API. Evaluation responses nest every enrichment result under `data_enrichments`, and the only size control is `include_input`, which is off by default. The MCP server has nine compact tools, though none runs an evaluation (10 of 25). No endpoint lists or searches evaluations, so a caller needs the `eval_id`. The reason code catalogue filters by product, name, text and active status (6 of 20). Errors come with one of 15 codes, field-level `invalid_args` and a list of which statuses to retry, though some legacy endpoints omit `code` (18 of 20). No idempotency header is documented. Completed evaluations are immutable, webhook receivers can dedupe on `event_id`, and the MCP tool annotations could not be read without a key (6 of 20). A request needs only `id`, `timestamp`, `workflow` and `data`. The official SDKs are client-side capture and device libraries for web, iOS, Android and React Native, with no server-side SDK found (6 of 15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels 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.\n\n## 2. Payments \u0026 pricing, 35 out of 100, up to 8.1 more on the total\n\nWhy it scored 35: Read with the hosted rubric. No x402, MPP or L402 (0). The pricing page publishes four Socure Launch solutions at $0.80, $0.90, $1.00 and $1.30 per evaluation without a login. Enterprise prices come from sales (20). The sandbox is free and Launch accounts receive $1,000 in production credits a month. We could not read the sign-up flow to confirm that no card is asked for, so this line is scored short of full (15 of 20). A person signs up in a browser, and production stays at 0 requests a second until Socure activates it (0).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment 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.\n\nOpen-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.\n\n## 3. Security \u0026 auth, 60 out of 100, up to 7 more on the total\n\nWhy it scored 60: Bearer API keys are separate for sandbox and production and can be regenerated with an overlap period, and the MCP server takes its own key. Each key reaches every workflow with no scopes, and an additional key is requested from Support. Optional mutual TLS, an IP allowlist and JWE payload encryption add network and transport controls (22 of 30). No read-only key or per-endpoint permission exists. The MCP key cannot read results or run evaluations but can create and change webhooks, and dashboard roles include a Developer role with no access to personal data (8 of 20). Responses carry end-user input, document extractions and adverse media. The MCP page tells users to treat AI output as advisory and keep keys out of prompts, with no guidance on untrusted content in responses (4 of 15). The dashboard has audit logs for account and workflow changes and CSV compliance reports. No API request log was found (9 of 15). SOC 2 Type 2, ISO/IEC 27001:2022, 27017, 27018 and 27701, FedRAMP Moderate, and a disclosure programme hosted by Bugcrowd. security.txt expired on 15 July 2026, and no public advisories or bounty terms were found (17 of 20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels 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.\n\n## 4. Reliability, 75 out of 100, up to 5 more on the total\n\nWhy it scored 75: Read with the hosted lines and scored on the Evaluation API. status.socure.com is a Statuspage site with RiskOS and each product as components and history back to December 2025 (20). It lists 13 incidents between 14 July and 22 September 2026, all marked minor or none. Most are short latency or error-rate events on single products, such as 5xx errors on fraud, KYC and watchlist for 20 minutes on 9 September, and one is an eCBSV latency incident of about 21 hours on 2 and 3 September that Socure attributes to the Social Security Administration (20 of 30). Limits have numbers, 10 requests a second and 1,000 a day per endpoint and method in the sandbox, 1 evaluation a second and 500 a day on Socure Launch, and a per-account figure in production (15). 429 responses carry `X-Retry-After`, and the docs give a backoff pattern with jitter. The same page advises idempotency keys, but no idempotency header is defined in the docs or the OpenAPI file (10 of 15). The trust centre says service levels are set in each customer's contract, and no SLA is published (0). The API is generally available under version 2025-01-01.orion (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 5. Transparency \u0026 trust, 68 out of 100, up to 2.8 more on the total\n\nMade of editorial 53, provenance 82.\n\nWhy it scored 68: The service is closed, and no service agreement or API terms are published. The terms page is the Website Terms of Use, the trust centre hides its terms entry, and the OpenAPI file calls the docs confidential while they are served publicly (5 of 30). The Global Services Privacy Notice, effective 1 October 2026, separates Socure's controller and processor roles and gives maximum retention periods, up to seven years for most categories and one to three years for others. It says documents and images may be used to train and test models, and the go-live checklist says customer retention is set by contract. No DPA is public, and the trust centre still cites the 19 June 2026 version of the notice (20 of 30). SDK versions have a written end-of-support policy with dates, and the removal of password sign-in for the MCP server on 21 July 2026 and new source addresses after 31 August 2026 were announced with dates. No deprecation policy for API versions was found (13 of 20). The trust centre names Amazon Web Services in US East, Google Cloud for RiskOS and Snowflake, all in the United States, and the privacy notice adds service providers in the Philippines and staff access from India, the United Kingdom and Europe (15 of 20).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\nProvenance checks not met in full (half of this category, computed from checked facts):\n\n- Terms of service: not found (0 of 10)\n- Privacy policy: read, states 7 of the 8 things a reader expects, and has 1 clause that costs points (7.3 of 10)\n- security.txt: published but past its Expires date (5 of 10)\n\n## 6. Schema \u0026 documentation, 84 out of 100, up to 2.6 more on the total\n\nWhy it scored 84: An OpenAPI 3.0.3 file for ten operations and a 3.1 file for the webhook payload are public on GitHub and embedded in each reference page (25). llms.txt for the site, the project and each section, and every page as Markdown by adding `.md` (10). Operation descriptions are short, from 76 to 689 characters, while the guides for each solution say when to choose it and which integration path it supports (12 of 20). The file has 39 enums and 47 required lists, but the evaluation request body declares no required fields at the top level, 20 schemas allow arbitrary properties, and the request and response schemas for each enrichment are in the dashboard, not the public file. Reason code and signal documentation moved to the dashboard in February 2026 (8 of 15). The file carries 357 example values, each operation documents its 4xx and 5xx responses, and the errors page lists 15 codes with sample bodies (14 of 15). An `X-API-Version` header pins the version, and dated release notes are public (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels 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.\n\n## 7. Maintenance \u0026 community, 75 out of 100, up to 2.2 more on the total\n\nWhy it scored 75: The latest release note is dated 6 October 2026 (30). More than 20 dated entries since 10 July 2026, among them 5 October, 29 September and 22 September (20). A closed service with public release notes and an email subscription, email support, chat for Launch accounts and a support department the go-live checklist describes as 24/7. We did not test a channel (10 of 15). `@socure-inc/device-risk-sdk` reached 2.11.0 on 1 September 2026 and the DocV SDKs are at v5, all client-side. No server-side SDK and no MCP registry entry were found (10 of 15). The public spec repository was synced on 21 separate days since 10 July, most recently on 5 October. No CI is visible, and the integration skill repository was last changed on 23 February 2026 (5 of 10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## What we couldn't check\n\nWhat 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.\n\n- unchecked: the sign-up flow at riskos.sandbox.socure.com, whose robots.txt disallows every path, so whether a card or business verification is asked for at sign-up or before production is not established\n- unchecked: the MCP server's tool definitions, input schemas and annotations, which need an MCP Server Key\n- unchecked: the Enrichment API Reference, reason codes and signal definitions, which are in the dashboard\n- unchecked: GitHub star counts and any SDK source repositories beyond the two cloned\n- No service agreement, API terms, DPA or SLA is published, so `provenance.terms` is left out\n- The Launch FAQ says the sandbox has no limit, while the rate limit page gives 1,000 requests a day\n- The lead described the MCP server as an interface to the product. It reads docs and account configuration and manages webhooks, and cannot run or read evaluations\n- The Website Terms of Use forbid robots or other automatic software from monitoring or copying the website without written permission, while robots.txt on both sites allows crawling. Recorded as a fact\n- Socure also runs ID+ and a separate government cloud, each with its own docs. Neither was graded here\n\n## Weaknesses\n\n- No service agreement or API terms are published. The only terms pages are website terms of use and end-user terms for document capture\n- Each API key reaches every workflow, with no scopes or read-only keys, and an additional key has to be requested from Support\n- The rate limit page advises idempotency keys, but neither the docs nor the OpenAPI file define an idempotency header\n- No endpoint lists or searches evaluations. A caller needs the `eval_id` from the original response or a webhook\n- Enrichment request and response schemas, reason codes and signal definitions sit in the dashboard, not in the public docs\n- Production starts at 0 requests a second until Socure approves it, and Launch accounts are capped at 1 evaluation a second and 500 a day\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Send `Authorization: Bearer \u003ckey\u003e` to `https://riskos.sandbox.socure.com/api/evaluation` for tests and `https://riskos.socure.com/api/evaluation` for live checks. The two environments use separate keys.\n- Use dummy data only in the sandbox. It returns predefined results, connects to no live data source and expects the documented test values.\n- Store the `eval_id` from every `POST /api/evaluation` response. No list or search endpoint exists to find it later.\n- On 429 wait the seconds given in `X-Retry-After`. Do not blindly resend a failed `POST /api/evaluation`, because no idempotency header is documented and Launch bills each initiated evaluation.\n- The MCP server at `https://mcp.riskos.socure.com/sandbox` reads docs, workflows and webhooks with its own MCP Server Key. It cannot run evaluations or read results.\n\n## When it's done\n\nSend 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.\n",
    "grade": "B",
    "score": 63.5,
    "assessed": "2026-10-08",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 46,
        "maxGain": 8.8,
        "reason": "Scored on the REST API. Evaluation responses nest every enrichment result under `data_enrichments`, and the only size control is `include_input`, which is off by default. The MCP server has nine compact tools, though none runs an evaluation (10 of 25). No endpoint lists or searches evaluations, so a caller needs the `eval_id`. The reason code catalogue filters by product, name, text and active status (6 of 20). Errors come with one of 15 codes, field-level `invalid_args` and a list of which statuses to retry, though some legacy endpoints omit `code` (18 of 20). No idempotency header is documented. Completed evaluations are immutable, webhook receivers can dedupe on `event_id`, and the MCP tool annotations could not be read without a key (6 of 20). A request needs only `id`, `timestamp`, `workflow` and `data`. The official SDKs are client-side capture and device libraries for web, iOS, Android and React Native, with no server-side SDK found (6 of 15).",
        "checklist": [
          "- 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).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 35,
        "maxGain": 8.1,
        "reason": "Read with the hosted rubric. No x402, MPP or L402 (0). The pricing page publishes four Socure Launch solutions at $0.80, $0.90, $1.00 and $1.30 per evaluation without a login. Enterprise prices come from sales (20). The sandbox is free and Launch accounts receive $1,000 in production credits a month. We could not read the sign-up flow to confirm that no card is asked for, so this line is scored short of full (15 of 20). A person signs up in a browser, and production stays at 0 requests a second until Socure activates it (0).",
        "checklist": [
          "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.\n- 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.\n- 20, a free tier or trial that doesn't need a card.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 60,
        "maxGain": 7,
        "reason": "Bearer API keys are separate for sandbox and production and can be regenerated with an overlap period, and the MCP server takes its own key. Each key reaches every workflow with no scopes, and an additional key is requested from Support. Optional mutual TLS, an IP allowlist and JWE payload encryption add network and transport controls (22 of 30). No read-only key or per-endpoint permission exists. The MCP key cannot read results or run evaluations but can create and change webhooks, and dashboard roles include a Developer role with no access to personal data (8 of 20). Responses carry end-user input, document extractions and adverse media. The MCP page tells users to treat AI output as advisory and keep keys out of prompts, with no guidance on untrusted content in responses (4 of 15). The dashboard has audit logs for account and workflow changes and CSV compliance reports. No API request log was found (9 of 15). SOC 2 Type 2, ISO/IEC 27001:2022, 27017, 27018 and 27701, FedRAMP Moderate, and a disclosure programme hosted by Bugcrowd. security.txt expired on 15 July 2026, and no public advisories or bounty terms were found (17 of 20).",
        "checklist": [
          "- 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.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 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.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 75,
        "maxGain": 5,
        "reason": "Read with the hosted lines and scored on the Evaluation API. status.socure.com is a Statuspage site with RiskOS and each product as components and history back to December 2025 (20). It lists 13 incidents between 14 July and 22 September 2026, all marked minor or none. Most are short latency or error-rate events on single products, such as 5xx errors on fraud, KYC and watchlist for 20 minutes on 9 September, and one is an eCBSV latency incident of about 21 hours on 2 and 3 September that Socure attributes to the Social Security Administration (20 of 30). Limits have numbers, 10 requests a second and 1,000 a day per endpoint and method in the sandbox, 1 evaluation a second and 500 a day on Socure Launch, and a per-account figure in production (15). 429 responses carry `X-Retry-After`, and the docs give a backoff pattern with jitter. The same page advises idempotency keys, but no idempotency header is defined in the docs or the OpenAPI file (10 of 15). The trust centre says service levels are set in each customer's contract, and no SLA is published (0). The API is generally available under version 2025-01-01.orion (10).",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 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.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 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.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 68,
        "maxGain": 2.8,
        "reason": "The service is closed, and no service agreement or API terms are published. The terms page is the Website Terms of Use, the trust centre hides its terms entry, and the OpenAPI file calls the docs confidential while they are served publicly (5 of 30). The Global Services Privacy Notice, effective 1 October 2026, separates Socure's controller and processor roles and gives maximum retention periods, up to seven years for most categories and one to three years for others. It says documents and images may be used to train and test models, and the go-live checklist says customer retention is set by contract. No DPA is public, and the trust centre still cites the 19 June 2026 version of the notice (20 of 30). SDK versions have a written end-of-support policy with dates, and the removal of password sign-in for the MCP server on 21 July 2026 and new source addresses after 31 August 2026 were announced with dates. No deprecation policy for API versions was found (13 of 20). The trust centre names Amazon Web Services in US East, Google Cloud for RiskOS and Snowflake, all in the United States, and the privacy notice adds service providers in the Philippines and staff access from India, the United Kingdom and Europe (15 of 20).",
        "blend": "editorial 53, provenance 82",
        "checklist": [
          "- 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.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 84,
        "maxGain": 2.6,
        "reason": "An OpenAPI 3.0.3 file for ten operations and a 3.1 file for the webhook payload are public on GitHub and embedded in each reference page (25). llms.txt for the site, the project and each section, and every page as Markdown by adding `.md` (10). Operation descriptions are short, from 76 to 689 characters, while the guides for each solution say when to choose it and which integration path it supports (12 of 20). The file has 39 enums and 47 required lists, but the evaluation request body declares no required fields at the top level, 20 schemas allow arbitrary properties, and the request and response schemas for each enrichment are in the dashboard, not the public file. Reason code and signal documentation moved to the dashboard in February 2026 (8 of 15). The file carries 357 example values, each operation documents its 4xx and 5xx responses, and the errors page lists 15 codes with sample bodies (14 of 15). An `X-API-Version` header pins the version, and dated release notes are public (15).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 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.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 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."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 75,
        "maxGain": 2.2,
        "reason": "The latest release note is dated 6 October 2026 (30). More than 20 dated entries since 10 July 2026, among them 5 October, 29 September and 22 September (20). A closed service with public release notes and an email subscription, email support, chat for Launch accounts and a support department the go-live checklist describes as 24/7. We did not test a channel (10 of 15). `@socure-inc/device-risk-sdk` reached 2.11.0 on 1 September 2026 and the DocV SDKs are at v5, all client-side. No server-side SDK and no MCP registry entry were found (10 of 15). The public spec repository was synced on 21 separate days since 10 July, most recently on 5 October. No CI is visible, and the integration skill repository was last changed on 23 February 2026 (5 of 10).",
        "checklist": [
          "- 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.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 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.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      }
    ],
    "provenance": [
      {
        "label": "Terms of service",
        "value": "not found",
        "points": 0,
        "max": 10
      },
      {
        "label": "Privacy policy",
        "value": "read, states 7 of the 8 things a reader expects, and has 1 clause that costs points",
        "points": 7.3,
        "max": 10
      },
      {
        "label": "security.txt",
        "value": "published but past its Expires date",
        "points": 5,
        "max": 10
      }
    ],
    "unchecked": [
      "unchecked: the sign-up flow at riskos.sandbox.socure.com, whose robots.txt disallows every path, so whether a card or business verification is asked for at sign-up or before production is not established",
      "unchecked: the MCP server's tool definitions, input schemas and annotations, which need an MCP Server Key",
      "unchecked: the Enrichment API Reference, reason codes and signal definitions, which are in the dashboard",
      "unchecked: GitHub star counts and any SDK source repositories beyond the two cloned",
      "No service agreement, API terms, DPA or SLA is published, so `provenance.terms` is left out",
      "The Launch FAQ says the sandbox has no limit, while the rate limit page gives 1,000 requests a day",
      "The lead described the MCP server as an interface to the product. It reads docs and account configuration and manages webhooks, and cannot run or read evaluations",
      "The Website Terms of Use forbid robots or other automatic software from monitoring or copying the website without written permission, while robots.txt on both sites allows crawling. Recorded as a fact",
      "Socure also runs ID+ and a separate government cloud, each with its own docs. Neither was graded here"
    ],
    "weaknesses": [
      "No service agreement or API terms are published. The only terms pages are website terms of use and end-user terms for document capture",
      "Each API key reaches every workflow, with no scopes or read-only keys, and an additional key has to be requested from Support",
      "The rate limit page advises idempotency keys, but neither the docs nor the OpenAPI file define an idempotency header",
      "No endpoint lists or searches evaluations. A caller needs the `eval_id` from the original response or a webhook",
      "Enrichment request and response schemas, reason codes and signal definitions sit in the dashboard, not in the public docs",
      "Production starts at 0 requests a second until Socure approves it, and Launch accounts are capped at 1 evaluation a second and 500 a day"
    ],
    "agentNotes": [
      "Send `Authorization: Bearer \u003ckey\u003e` to `https://riskos.sandbox.socure.com/api/evaluation` for tests and `https://riskos.socure.com/api/evaluation` for live checks. The two environments use separate keys.",
      "Use dummy data only in the sandbox. It returns predefined results, connects to no live data source and expects the documented test values.",
      "Store the `eval_id` from every `POST /api/evaluation` response. No list or search endpoint exists to find it later.",
      "On 429 wait the seconds given in `X-Retry-After`. Do not blindly resend a failed `POST /api/evaluation`, because no idempotency header is documented and Launch bills each initiated evaluation.",
      "The MCP server at `https://mcp.riskos.socure.com/sandbox` reads docs, workflows and webhooks with its own MCP Server Key. It cannot run evaluations or read results."
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-09",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.4",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}