Weaviate API + MCP by Weaviate

# Fix list: Weaviate API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/weaviate, the October 2026 research run, assessed 1 October 2026. Grade B, 68.2 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 Weaviate API + MCP: 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. Reliability, 35 out of 100, up to 13 more on the total

Why it scored 35: No public status page. status.weaviate.io doesn't resolve, and status.weaviate.cloud is titled "Weaviate Enterprise Status" with a page-not-found body and an empty incident feed (0). No readable history (5). No request limits published for the database. The only numbers are on the AI services (Embeddings 2,000 requests a day and Query Agent 1,000 a month on Free) and a 60-second request timeout on Weaviate Cloud (5 of 15). No 429 or backoff guidance found, but the import docs tell you to use deterministic UUIDs so a retried write can't duplicate objects (5 of 15). SLA of 99.5% on Flex, 99.9% on Premium shared and 99.95% on dedicated (10). The database API is GA and the docs page for the built-in MCP server carries an "Added in v1.38" note with no preview label (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.

## 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). Per-unit prices published without login, from $0.00465 per million vector dimensions and $0.12 per GiB stored on Flex, with Premium rates listed too (20). Free cluster with no card required, 100,000 objects and 1 collection (20). A person signs up in the browser and creates a key per cluster (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. Security & auth, 79 out of 100, up to 3.7 more on the total

Why it scored 79: Weaviate Cloud keys are created per cluster with an RBAC role (admin, viewer or custom), and can be rotated and deleted. Self-hosted instances add OIDC (30). Viewer roles, write tools off by default on self-hosted MCP and a read-only toggle on Weaviate Cloud, though that toggle is off by default (18). Search returns stored objects as written and the MCP docs give no prompt-injection guidance (5). With RBAC on, authorisation decisions and role changes are written to an audit log, with source IP. Not a full per-call log (10). SOC 2 Type II and HIPAA on the pricing page, a trust portal at trust.weaviate.io and a vulnerability report form. No SECURITY.md and no security.txt, and only one advisory ever published, in 2023 (16).

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.

## 4. Schema & documentation, 82 out of 100, up to 2.9 more on the total

Why it scored 82: Swagger 2.0 file at docs.weaviate.io/openapi.json covers REST only, and search and bulk import run over gRPC. All 4 MCP tools have typed JSON Schema inputs (22). llms.txt (10). MCP descriptions are one line each, such as "Performs hybrid search (vector + keyword) for data in a collection.", with nothing on when to use or avoid a tool. Operators can override them in a config file (10). Typed inputs with required fields, an operator enum in the filter schema and `alpha` documented as 0 for keyword and 1 for vector with a 0.75 default (13). Code examples on most docs pages, error responses documented unevenly (12). Semver releases and dated release notes (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. Transparency & trust, 71 out of 100, up to 2.5 more on the total

Made of editorial 69, provenance 72.

Why it scored 71: BSD-3-Clause for the core. Since 26 August 2026 a `wl/` directory holds enterprise code (namespaces, self-recovery) under a proprietary licence unlocked by key, so the repository is open-core now (25). The privacy policy, amended March 2026, names Weaviate B.V. in Amsterdam, lists processors in section 8, states a 5-year retention period and the transfer basis, and agrees with the telemetry page. No DPA link found (22). Deprecations are flagged per setting in the docs with the version, but there's no notice-period policy (10). Self-hosted telemetry is on by default and documented with `DISABLE_TELEMETRY`, but the docs say Weaviate may use it to find sales prospects and share it with third parties (12).

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: weaviate.io, registered 2022-01-30 (4 years) (7 of 15)
- Status page: not found (0 of 10)
- security.txt: not found (0 of 10)

## 6. Agent ergonomics, 92 out of 100, up to 1.3 more on the total

Why it scored 92: 4 MCP tools (25). `limit`, `return_properties`, `return_metadata`, `target_properties`, filters and tenants on hybrid search, plus a cursor API for paging through a collection (20). Errors come back as JSON messages, and with write access off the upsert tool returns a clear write-disabled error instead of "tool not found". No error-code catalogue (12). Every tool sets readOnlyHint, destructiveHint and idempotentHint, and write tools are hidden from tools/list until write access is enabled (20). Only `query` and `collection_name` are required. Official clients for Python, TypeScript, Go, Java and C# (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. Maintenance & community, 90 out of 100, up to 0.9 more on the total

Why it scored 90: v1.39.8 tagged on 1 October 2026 (30). Dozens of patch releases across three supported minors since July, with v1.39.0 to v1.39.8 alone in two months (20). 456 open issues, the newest bug reports from early September, and we couldn't see reply counts (15). Official clients in five languages (15). CI on pull requests with a linter and a security lint workflow (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.

- Whether Weaviate Cloud has a public status page anywhere other than status.weaviate.io and status.weaviate.cloud
- Whether the built-in MCP server is still considered preview. The listing said so on 30 September, and the docs page we read on 1 October carries no preview label
- How quickly maintainers reply to bug reports, since the issue list didn't show reply counts

## Weaknesses

- No public status page or incident history
- No request limits or 429 guidance published for the database API
- Self-hosted telemetry is on by default, and the docs say it may be shared with third parties to find customers
- The OpenAPI file covers REST only. Search and bulk import lean on gRPC and the client libraries
- On Weaviate Cloud the MCP endpoint is always on and allows writes until someone flips the read-only toggle

## 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.

- Call `weaviate-collections-get-config` first so the model sees property names before it writes a filter
- `weaviate-query-hybrid` defaults `alpha` to 0.75, so lower it for keyword-heavy queries
- Turn on the read-only toggle in Weaviate Cloud before giving an agent the MCP endpoint
- Set object IDs with `generate_uuid5` so a retried import doesn't create duplicates
- Free clusters hold one collection. Use tenants rather than extra collections

## What the review panel asked for

- a stated notice period
- a public status page
- Publish the full tier table

## 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.