Typesense API + MCP by Typesense

# Fix list: Typesense API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/typesense, the October 2026 research run, assessed 1 October 2026. Grade BB, 70.9 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 Typesense 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. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: No x402, MPP or L402 (0). Hourly pricing by RAM, vCPU, nodes and region through a calculator, bandwidth per GB, no rate table on the page (10). A free-tier cluster (0.5 GB RAM, 2 vCPU burst, one node) needs no payment method (20). A person signs up in the browser. A management key lets a headless agent provision clusters over MCP, but a person creates that 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.

## 2. Security & auth, 58 out of 100, up to 7.4 more on the total

Why it scored 58: Cluster keys can be limited to collections and actions with an expiry, scoped search keys embed filters, management keys carry named capabilities and the hosted MCP server uses OAuth with consent presets, which would be 30. Less 10 because the authentication docs show the key in the URL as `x-typesense-api-key=` in a curl example (20). Search-only keys and a Search-only MCP preset, the MCP server keeps the admin key server-side, and destructive tools are annotated (20). Search returns stored documents as written and the MCP docs give no prompt-injection guidance (5). Cluster configuration and cloning history is visible through `get_clusters`, but we found no per-call audit log (3). SECURITY.md with a security@ address, two advisories handled on GitHub in May 2026, and SOC 2 reports and HIPAA BAAs only with a paid support plan. No security.txt per the 30 September check and no bounty found (10).

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.

## 3. Reliability, 83 out of 100, up to 3.4 more on the total

Why it scored 83: Instatus page at cloud-status.typesense.org with the management console and 21 cluster regions as components (20). No notices from July to October 2026. The last incident was 11 hours 30 minutes of slow new-cluster provisioning on 31 May, with existing clusters unaffected (30). Limits documented with numbers, 300 data calls and 30 cluster actions a minute per MCP connection, 30 a minute per management key, and the docs state there are no HTTP rate limits on the server API (15). The management API returns 429 with "Please try again later", with no `Retry-After` or backoff advice. Imports take an `upsert` action, so a retried import is safe (8 of 15). No SLA found on the pricing page or in the docs we read (0). GA (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. Maintenance & community, 72 out of 100, up to 2.5 more on the total

Why it scored 72: typesense-js 3.1.0 on 25 September 2026 (30). The server's last stable release is v30.2, tagged 9 April 2026. Since July we found typesense-js 3.1.0, the hosted MCP launch and an OpenAPI update, but no three releases (10). The v31 branch takes commits daily, but 805 issues are open and many new ones sit at pre-triage (12). Official clients current with v30 (15). CI runs tests on the server, but an open issue reports the bundled OpenSSL pinned at 3.0.5, a branch past end of life (5).

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.

## 5. Schema & documentation, 86 out of 100, up to 2.3 more on the total

Why it scored 86: OpenAPI 3.0.3 in typesense/typesense-api-spec, 79 operations, version 30.0, last updated 10 September 2026 (25). llms.txt and Markdown docs (10). The MCP docs say `whoami` should be called first and describe each tool, but we couldn't read the hosted tool definitions themselves (14). The server API is typed in the spec, but two MCP tools, `read_cluster_data` and `change_cluster_data`, take any server API path, which leaves the shape to the model (10). An API errors page lists codes 400 to 503, and every docs page has examples in several languages (12). Docs versioned per server release, GitHub releases and a versioned spec (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, 80 out of 100, up to 1.8 more on the total

Made of editorial 74, provenance 86.

Why it scored 80: GPL-3.0 (30). The Typesense Cloud privacy policy dates from 20 September 2022, names Typesense Inc., lists processors and AWS regions and links a DPA, but gives no retention period and says the full processor list is on request (18). Docs are versioned per release with an upgrade guide, but we found no deprecation notice policy (8). No usage telemetry in the server source that we could find, and hosted data locations are listed (18).

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: typesense.org, registered 2017-04-14 (9 years) (11 of 15)
- security.txt: not found (0 of 10)

## 7. Agent ergonomics, 93 out of 100, up to 1.1 more on the total

Why it scored 93: 9 MCP tools per the docs, including two generic read and change tools that cover the rest of the server API (25). `per_page`, `page`, `include_fields`, `exclude_fields`, `filter_by` and facets (20). A refused MCP call names the permission it lacks, and imports over 1,000 documents or 10 MB come back as a curl command with a 24-hour key scoped to one collection and action (18). `manage_cluster`, `documents` and `change_cluster_data` are marked destructive so clients ask first. We couldn't confirm readOnlyHint on the read tools (15). Official clients for JavaScript, Python, PHP, Ruby and more. The data tools need a browser OAuth sign-in, and a headless agent only gets the cluster tools (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.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- GHSA-97x4-gm45-jpcw (moderate, search cache not isolated between scoped search keys) and GHSA-fpx5-8c99-247j (high, unauthenticated denial of service on /multi_search), both published 18 May 2026. The cache-keying fix landed on 20 March 2026 before v30.2. Fixed and disclosed, so a small deduction (https://github.com/typesense/typesense/security).

## 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: the hosted MCP tool definitions themselves, including whether read tools set readOnlyHint
- Whether Typesense Cloud has an SLA on any plan
- When v31 will ship as a stable release

## Weaknesses

- Server v30.2 from April 2026 is still the newest stable release
- The authentication docs show the API key as a URL query parameter
- No SLA found, and SOC 2 reports need a paid support plan
- The MCP data tools need a browser sign-in, so a headless agent gets cluster tools only
- Cloud billing is per cluster-hour on dedicated hardware, so an idle cluster still costs money

## 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 `whoami` first on the MCP server to see which permissions the connection holds
- Put the embedding field in `query_by` alongside text fields for hybrid search
- Send the key in `X-TYPESENSE-API-KEY`, not in the URL, so it stays out of logs
- Use a scoped search key for retrieval-only agents; never pass the admin key
- Imports over 1,000 documents or 10 MB through MCP come back as a curl command to run locally

## What the review panel asked for

- a v31 release date
- a deprecation policy
- Publish the hourly rate 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.