Melius

# Fix list: Melius

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/melius, the October 2026 research run, assessed 8 October 2026. Grade C, 54.1 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 Melius: 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, 43 out of 100, up to 10 more on the total

Why it scored 43: API keys are revocable at once, can expire and are shown once, but have no scopes. A key works with its creator's access in every team that user belongs to, capped at the user's role. The MCP OAuth flow uses S256 PKCE and dynamic client registration, lists no scopes, and ties the connector key to one team (22 of 30). A key made by a viewer can only read, `bulk_delete_node` asks the user to confirm and a run can return a pending approval, but there is no read-only key and the setup guide tells users to select Always Allow for every tool (9 of 20). Canvas text, comments from collaborators and files fetched by `url_to_canvas` reach the model, and no prompt-injection guidance was found (2 of 15). Keys list a `lastUsedAt` time, connector keys are named after the client, and generation history is kept per member. No audit log is documented (7 of 15). No security.txt, disclosure policy or bug bounty found. A Vanta trust centre exists at trust.melius.com but needs JavaScript, so certifications are unchecked (3 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, 20 out of 100, up to 10 more on the total

Why it scored 20: No machine payment protocol (0). Plan prices and monthly credits are public, but per-model credit costs sit behind an authenticated endpoint and the top-up price is not published (10 of 20). A Free plan at $0 gives trial credits. The amount is not published and we couldn't confirm that sign-up needs no card (10 of 20). A person signs up in a browser and creates the first key in Team settings. POST /api-keys needs an existing key (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.

## 3. Agent ergonomics, 51 out of 100, up to 8 more on the total

Why it scored 51: The MCP reference lists 75 tools, in the over-30 band, with `get_guide` and `get_docs` loading guidance on demand and 18 Cast tools shown only to teams with access (10 of 25). Four list operations take `limit` and `offset` and three take `search`, canvas content can be narrowed to one node or to media summaries, and the CLI has `--fields` (14 of 20). REST errors carry `statusCode`, `error` and `message`, the CLI adds a `code` and a `suggestion`, and `CANVAS_TOO_LARGE` is documented as not worth retrying. A shortage of credits arrives as a plain 400 (14 of 20). No idempotency keys, and we couldn't read MCP annotations without an account. The docs do give safe-retry guidance (5 of 20). Auto model selection and defaults keep required fields few. There is no SDK library, only the CLI and generated code samples in six languages (8 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.

## 4. Reliability, 69 out of 100, up to 6.2 more on the total

Why it scored 69: Graded on the hosted REST API and MCP server. status.melius.com runs on incident.io with component history for Web app and MCP, though it has no API component (20). Its history lists no incidents from July to October 2026, which is as far back as the page goes (30). No rate-limit numbers are published. The docs say only that no plan has a monthly request quota and that bursts can return 429 (0). 429s carry `Retry-After` and a `retryAfterSeconds` body field, and the docs give a bounded retry policy and tell clients to check canvas or run state after a timeout or 5xx on a write. There are no idempotency keys (12 of 15). No SLA found (0). The API and MCP server carry no beta label, but the CLI is at 0.16.2 and Cast and media-URL inputs are marked beta (7 of 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. Schema & documentation, 70 out of 100, up to 4.9 more on the total

Why it scored 70: Public OpenAPI 3.0.2 spec with 59 operations (25). llms.txt, llms-full.txt and a Markdown copy of every docs page (10). 58 of the 59 operations have a summary and no description. The MCP reference gives each of 75 tools one line, with ordering rules for `get_guide`, `show_presence` and `canvas_plan_layout`, and the tool definitions themselves need a signed-in client, so we didn't read them (8 of 20). Request bodies are typed with 336 enums, ranges, UUID formats and required fields, with 52 open `additionalProperties` objects (13 of 15). 401, 403 and 429 are declared on every operation and 400, 404 and 409 where they apply, but the spec has no examples and most error bodies are a bare `message` (9 of 15). The path is versioned at /api/v1 and the CLI states semantic versioning. No public changelog found (5 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.

## 6. Transparency & trust, 55 out of 100, up to 3.9 more on the total

Made of editorial 38, provenance 72.

Why it scored 55: Closed service under published terms effective 28 August 2026. The CLI package is MIT with no public repository (15 of 30). The terms, the privacy policy and the pricing FAQ agree that user input and outputs are not used to train models, and say third-party model providers process content under their own terms without a guarantee from Melius. Retention is described as a reasonable period with no numbers, and no DPA was found (14 of 30). Deprecated models keep working for an unstated grace period with an in-app warning. No API deprecation policy found (5 of 20). No sub-processor list or hosting location found beyond a statement that data is processed in the United States (4 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, and has 3 clauses that cost points (2.3 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects (9.3 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 66 out of 100, up to 3 more on the total

Why it scored 66: The newest dated release is the CLI, 0.16.2 on 10 September 2026, 28 days before the check (30). npm lists 19 CLI versions since 10 July 2026 (20). Closed service with no public changelog or issue tracker. Support is by email at support@melius.com and a Discord server linked from the site (5 of 15). The server is not in the official MCP registry and there is no SDK library, though the CLI is current on npm and Homebrew (5 of 15). The CLI has three dependencies, needs Node.js 20 and is published from GitHub Actions through npm trusted publishing. Its source is not public (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: trust.melius.com is a Vanta trust centre that needs JavaScript, so certifications, sub-processors and security policies listed there were not read.
- unchecked: MCP tool definitions and annotations (readOnlyHint, destructiveHint), which need a signed-in client.
- unchecked: whether the Free plan needs a card, how many trial credits it gives, and whether Free teams can create API keys. The sign-up flow is at app.melius.com behind a login.
- The docs say the `mel-mcp` stdio binary ships with the CLI, but @melius-ai/cli 0.16.2 declares only `mel` and its bundle has no MCP command. We found no other package that carries it.
- The pricing page shows two sets of figures. Plan cards give $20, $50, $110 and $70 monthly ($17, $43, $93, $56 on annual billing), and the comparison table gives $18, $45, $99+ and $63.
- No rate-limit numbers, MCP usage limits, SLA, changelog or API deprecation policy were found. The docs mention a team MCP usage record without describing a cap.
- Category kept as design with `design.canvas` first, because every generation is a canvas node and the API has no separate generate endpoint. Image generation is the alternative.

## Weaknesses

- API keys have no scopes. A key works with its creator's role in every team the creator belongs to
- No rate-limit numbers, SLA, public changelog or API deprecation policy found in the reviewed documentation
- The MCP reference lists 75 tools, and the setup guide tells users to select Always Allow for all of them
- Per-model credit costs are returned by an authenticated endpoint and shown in the app, not on the pricing page
- No security.txt or disclosure policy found, and the Vanta trust centre at trust.melius.com needs JavaScript we couldn't run

## 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.

- Generate by creating a node on a canvas, starting a run with POST /nodes/{nodeId}/runs, then polling GET /node-runs/{nodeRunId} until `status` is finished or failed
- Over MCP, call `get_guide` first, `show_presence` before node changes and `canvas_plan_layout` before `bulk_create_nodes`
- On 429 wait `Retry-After` seconds. After a timeout or 5xx on a write, read the canvas or run state before resubmitting, because there are no idempotency keys
- Read credit costs from GET /generation/models?category=image before a bulk run. A 400 can mean the team is out of credits
- Downloads arrive as a ZIP from a signed URL. Fetch that URL without the `Authorization` header

## 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.