Akeyless (SecretlessAI and MCP server) by Akeyless

# Fix list: Akeyless (SecretlessAI and MCP server)

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/akeyless, the October 2026 research run, assessed 1 October 2026. Grade BB, 73.7 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 Akeyless (SecretlessAI and MCP server): 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, 25 out of 100, up to 9.4 more on the total

Why it scored 25: No x402, MPP or L402 (0). The free plan's limits are public and everything above it is quoted with no figures (5 of 20). A free plan with no time limit. The pricing page doesn't mention a card, and the account quickstart lists the sign-up steps as email, terms, a reCAPTCHA, email verification, a password and profile details, with no payment step (20). A person signs up in a browser and creates the first access key or auth method (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. Agent ergonomics, 63 out of 100, up to 6 more on the total

Why it scored 63: akeyless mcp-runtime-authority has 4 tools, and akeyless mcp lists 9 and says there are more, with no full count published (18 of 25). `/list-items` takes a pagination token and filters by path, type, tag and modification date, with a `minimal-view` flag, per the OpenAPI document (20). Errors come back as one `error` string with no code for an agent to branch on (8 of 20). No tool annotations or idempotency guidance found, though the runtime-authority tools return results rather than secrets (5 of 20). SDKs for Python, Go, Node, Ruby and more, but every call needs a token in the body and many parameters (12 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.

## 3. Schema & documentation, 81 out of 100, up to 3.1 more on the total

Why it scored 81: The OpenAPI 3.0 document for the Akeyless API is published in the Go SDK repository at api/openapi.yaml, 657 POST paths (25). llms.txt at docs.akeyless.io with about 800 Markdown links, 300 of them API reference pages (10). The MCP page says what each of the two servers is for and which one to prefer when an agent shouldn't see credentials, but per-tool descriptions aren't published (12 of 20). The spec types every field and marks required ones, though the API is POST-only and the auth token travels in the JSON body of every call (10 of 15). Every path in the spec declares a default JSONError response, one `error` string with no code, and the schemas carry examples (9 of 15). Dated CLI changelog at changelog.akeyless.io, with deprecations called out (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.

## 4. Transparency & trust, 73 out of 100, up to 2.4 more on the total

Made of editorial 71, provenance 75.

Why it scored 73: SDKs are Apache-2.0, while the CLI, the Gateway and the service are closed under a master services agreement (18 of 30). The privacy policy (5 May 2026), a DPA (18 March 2026) and a sub-processor list (19 April 2026) are public. The DPA deletes or returns customer data on termination but gives no period in days, and it names Twilio as a sub-processor while the list files Twilio among vendors that don't touch customer personal data (18 of 30). The CLI changelog dates deprecations, for example the Explicitly Provide Credentials target mode in 1.147.0 (15 of 20). The list names five sub-processors, Google Cloud (including Vertex for Gemini), Azure, Azure OpenAI, Splunk and AWS, each with a US or US and Europe location (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: akeyless.io, no registry record we could read (0 of 15)
- security.txt: not found (0 of 10)

## 5. Reliability, 90 out of 100, up to 2 more on the total

Why it scored 90: Atlassian Statuspage at status.akeyless.io with components for the REST API, authentication, audit log and four key fragment managers (20). Since 3 July 2026 the history shows one incident, degraded dynamic secret performance in us-east-2 on 23 September for 21 minutes, plus a scheduled maintenance window on 19 July, so trivial only (30). The SLA and support tiers page (23 October 2025) sets transaction caps per support tier, 200 a minute and 2,000 a day on Silver, 600 and 4,000 on Gold, 800 and 6,000 on Platinum (15). Overuse is let through and counted as an extra client rather than refused, which documents what happens under load, but we found no Retry-After or backoff guidance (5 of 15). The same page commits to 99.99% availability on every tier, with service credits (10). The API is generally available and Runtime Authority went GA on 9 September 2026 (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.

## 6. Security & auth, 89 out of 100, up to 1.9 more on the total

Why it scored 89: Auth methods for AWS IAM, Azure AD, GCP, Kubernetes, OIDC, SAML, LDAP, JWT, certificates, Kerberos and Universal Identity, all mapped to RBAC, with the docs warning that access keys are for proofs of concept. The free plan excludes SAML, OIDC and LDAP auth (28 of 30). RBAC per path, a runtime-authority server that hands back query results, Runtime Authority intent rules with a kill switch, and CLI 1.151.0 added automatic locking on read for static and rotated secrets (18 of 20). SecretlessAI keeps the credential in the Gateway so the model never holds it (15). An audit log service on the status page, with 3 days of retention on the free plan (12 of 15). The trust centre lists SOC 2 Type II, ISO 27001 and ISO 27701 (certificates valid to 2028), PCI DSS and FIPS 140-3 validation (certificate 5227), and a bug bounty with a report page. No security.txt (404) and no public advisories page found (16 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.

## 7. Maintenance & community, 82 out of 100, up to 1.6 more on the total

Why it scored 82: CLI 1.152.0 on 16 September 2026, and the Python and Go SDKs at v5.0.38 on 17 September, 14 days before the run (30). Five CLI releases and nine SDK tags in the last 90 days (20). Public changelog, and the SLA page sets critical-issue response targets of 2 hours on Gold and 30 minutes on Platinum, with Silver best effort. We didn't test a reply (10 of 15). Official SDKs in several languages, current to 17 September 2026 (15). The Python SDK repository runs a test workflow and CodeQL, which we didn't see pass (7 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: an explicit statement that the free plan needs no card. The console sign-up page is closed to our fetcher by robots.txt, so the 20 points rest on the account quickstart's list of sign-up steps, which has no payment step
- unchecked: PyPI's current version. GitHub tags show v5.0.38 on 17 September 2026, which settles the 5.0.38 against 5.0.28 question in the listing's favour, but PyPI refused our fetcher on the re-check
- Whether calls over a tier's transaction cap ever get a 429, since the SLA page says overuse is allowed and counted as an extra client
- The DPA names Twilio as a sub-processor and the sub-processor list says it isn't one
- Per-tool definitions for both MCP servers, which aren't published

## Weaknesses

- No published prices above the free plan; clients and transactions are metered with overage billed at the end of a 12-month contract
- Errors come back as a single `error` string with no code, and no Retry-After or backoff guidance turned up
- The free plan has no OIDC, SAML or LDAP auth and keeps audit logs for 3 days
- akeyless mcp can return secret values to the model, and neither MCP server has a published tool list, version or registry entry
- No security.txt, and the closed CLI and Gateway are the only way to run the MCP servers

## 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.

- Run akeyless mcp-runtime-authority, not akeyless mcp, for an agent that acts on systems; the first returns results, the second has get_secret and get_password
- Authenticate with a cloud identity, Kubernetes or Universal Identity rather than an access key, which the docs reserve for proofs of concept
- Use CLI 1.130.0 or later for either MCP server, and point the client at your Gateway URL
- Count identities and calls before scaling. The free plan allows 5 clients, and calls over a tier's cap (200 a minute on Silver) are billed as extra clients, not refused
- Pass the token in the JSON body of each POST, and page `/list-items` with `pagination-token`

## What the review panel asked for

- versioned MCP servers
- a published MCP tool list
- publish a security.txt
- value-free mode for akeyless mcp

## 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.