# Fix list: AWS Secrets Manager From Anchor Terminal's listing at https://www.anchorterminal.com/tools/aws-secrets-manager, the October 2026 research run, assessed 1 October 2026. Grade A, 78.1 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 AWS Secrets Manager: 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, 20 out of 100, up to 10 more on the total Why it scored 20: No x402, MPP or L402 (0). $0.40 per secret a month and $0.05 per 10,000 API calls on the public pricing page (20). New customers since 15 July 2025 get up to $200 of Free Tier credit, but an AWS account takes a payment method at signup per the listing's 30 September check (0). A person creates the AWS account; there's no signup an agent can complete (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. Maintenance & community, 65 out of 100, up to 3.1 more on the total Why it scored 65: The newest change to the Secrets Manager API model in botocore is SortBy on ListSecrets (11 December 2025), before that managed external secrets (19 November 2025). AWS's open-source client for the service, the Workload Credentials Provider (formerly the Secrets Manager Agent), released 3.1.0 on 15 July and 3.1.1 on 21 July 2026, so we date the product by that (20). Two dated releases in the last 90 days that we could confirm (10 of 20). Public document history, AWS Support and re:Post, though the history page didn't load for us (10 of 15). Current official SDKs in every major language (15). SDKs release continuously and the Workload Credentials Provider ships a cargo-deny dependency policy and integration tests (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. ## 3. Reliability, 87 out of 100, up to 2.6 more on the total Why it scored 87: AWS Health Dashboard with per-service, per-region status and RSS feeds (20). The full dashboard history is a JavaScript page our fetcher can't read; the Secrets Manager us-east-1 feed had no items on 1 October 2026, so we can't confirm other regions and score partial (20 of 30). Quotas published per operation, GetSecretValue 10,000 a second, BatchGetSecretValue and ListSecrets 100, writes 50 (15). Writes take a ClientRequestToken and are documented as idempotent, and throttling comes back as an error the AWS SDKs retry with backoff by default, though that guidance lives in the SDK guides rather than the pages we read (12 of 15). SLA of 99.99% monthly uptime per region with 10, 25 and 100 per cent credits, last updated 5 December 2023 (10). 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. Security & auth, 88 out of 100, up to 2.1 more on the total Why it scored 88: IAM with SigV4, short-lived role credentials on EC2, ECS, Lambda and EKS, per-secret ARNs, condition keys and resource policies for cross-account grants; off AWS you can fall back to static access keys (30). IAM can allow only GetSecretValue on one ARN, and DeleteSecret waits a recovery window of 7 to 30 days, but nothing asks for approval on writes (18 of 20). The service returns no untrusted content (10). CloudTrail logs every call, including each GetSecretValue (15). A vulnerability disclosure programme on HackerOne and published security bulletins, but the security.txt at aws.amazon.com expired on 24 September 2026 and we didn't re-check certifications this run (15 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. Transparency & trust, 79 out of 100, up to 1.8 more on the total Made of editorial 63, provenance 95. Why it scored 79: Closed service under AWS Service Terms updated 15 September 2026, clear terms (15 of 30). Privacy notice of 18 May 2026 per the 30 September check, a DPA in the service terms, customer content stored in the regions you choose, and a recovery window of 7 to 30 days before a deleted secret is gone, with no retention schedule for request metadata found (22 of 30). No Secrets Manager deprecation policy or dated notices found (8 of 20). A sub-processor list updated 28 July 2026 that gives the processing location as the customer's selected region for most providers (18 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): - security.txt: published but past its Expires date (5 of 10) ## 6. Agent ergonomics, 90 out of 100, up to 1.6 more on the total Why it scored 90: GetSecretValue returns one secret, DescribeSecret returns metadata without the value, and BatchGetSecretValue takes a list of IDs or filters (20 of 25). ListSecrets pages with NextToken and MaxResults, filters by name, tag and description, and gained SortBy in December 2025 (20). Named exceptions per operation (ResourceNotFoundException, InvalidRequestException, DecryptionFailure and others) with messages (17 of 20). ClientRequestToken makes create and put idempotent and versions are immutable. There's no Secrets Manager MCP server to annotate; the general AWS API MCP server has a read-only mode and optional mutation consent (18 of 20). SDKs in every major language, GetSecretValue needs only a SecretId and defaults to AWSCURRENT, and AWS ships caching libraries plus the Workload Credentials Provider (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. ## 7. Schema & documentation, 96 out of 100, up to 0.7 more on the total Why it scored 96: Machine-readable service models (secretsmanager-2017-10-17) ship in every AWS SDK, with types, length limits, patterns and required members (25). llms.txt for the user guide with over 200 links to Markdown pages (10). The API reference says what each operation is for and when to hold back, such as caching GetSecretValue and not calling PutSecretValue more than once every 10 minutes (18 of 20). Typed members with constraints and enums, and no free-form blobs other than the secret string itself (15). Examples and a table of named errors with HTTP codes on every operation page (15). Dated API version and SDK changelogs; the user guide's document history page returned too many redirects for us (13 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. ## 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. - The listing had no llms.txt; docs.aws.amazon.com serves one for the Secrets Manager user guide, corrected in the patch. - The listing said AWS has no MCP route to Secrets Manager; the general AWS API MCP server can call it, corrected in notable and details. - unchecked: whether AWS signup still takes a card under the new Free Tier, relied on from the 30 September check's card-required tag. - unchecked: incident history outside us-east-1, since the Health Dashboard history is JavaScript-only. - The listing's lastRelease is empty. The API model last changed on 2025-12-11 and the official Workload Credentials Provider released on 2026-07-21; we left the field for the merge to decide. ## Weaknesses - Every API call is billed, so per-request reads add up and the docs push you to cache - Rotation outside the RDS family means writing and running a Lambda function - Off-AWS agents need AWS credentials of their own, often a static access key - No Secrets Manager MCP server, and the general AWS API server's read-only mode still allows GetSecretValue - The security.txt at aws.amazon.com expired on 24 September 2026, and signup needs a payment method ## 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. - Give the agent's task or instance role secretsmanager:GetSecretValue on the specific secret ARN, not a wildcard - Cache the value for the run, or read through the Workload Credentials Provider on localhost; each GetSecretValue is billed and logged - Use BatchGetSecretValue with a filter when you need several secrets at start-up; it's limited to 100 calls a second - Pass a ClientRequestToken on PutSecretValue so a retry can't create a second version, and don't write more than once every 10 minutes - Read VersionStage AWSPREVIOUS if a rotation lands mid-run and the new credential isn't live yet ## What the review panel asked for - Confirm card requirement - Value-masking MCP server - Managed rotation beyond RDS - State failed-call billing - fix document history page - Put retry guidance in the API reference - a published deprecation policy - value-free read-only MCP mode - renew the security.txt ## 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.