Upstash Vector API + MCP by Upstash

# Fix list: Upstash Vector API + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/upstash-vector, the October 2026 research run, assessed 1 October 2026. Grade B, 62.4 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 Upstash Vector 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. Schema & documentation, 49 out of 100, up to 8.3 more on the total

Why it scored 49: No OpenAPI file found for the Vector REST API. The hosted MCP server's 17 Vector and Search tools are closed, so we couldn't read their schemas (5 of 25). llms.txt and Markdown docs (10). Each REST endpoint page says what it does, with no when-not guidance (10). Typed request fields in the reference, but filters are a SQL-like string (10). curl and SDK examples on each page, and no error catalogue found (8). No API versioning, and the Vector changelog's last entry is August 2025 while the FAQ still says hybrid search isn't supported (6).

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.

## 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). $0.40 per 100,000 requests, $0.25 per GB stored and $0.03 per GB of bandwidth over 200 GB, published without login (20). Free plan with 10,000 queries and updates a day, no card per the 30 September check (20). A person signs up in the browser. The Developer API can create indexes, but only with an account key from a signed-up user (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, 63 out of 100, up to 6.5 more on the total

Why it scored 63: Each index has its own REST token and a separate read-only token, the Developer API uses account API keys with a read-only option, and the hosted MCP server defaults to OAuth with a per-client, revocable grant (28). Read-only tokens, read-only MCP grants and destructive labels on tools (20). Stored text comes back as written and we found no prompt-injection guidance (5). No audit log found (0). Valid security.txt per the 30 September check, a DPA, a subprocessor list and third-party pen tests, but SOC 2 covers Redis on Pro and Enterprise plans only and ISO 27001 is in progress (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.

## 4. Reliability, 75 out of 100, up to 5 more on the total

Why it scored 75: Statuspage at status.upstash.com with per-product components and an RSS history back to December 2025 (20). One Vector incident since July, about 20 minutes of intermittent DNS resolution errors in us-east-1 on 9 September 2026 with existing connections unaffected, so minor only (20). Daily caps published per plan (10,000 queries and updates on Free, 1M on the $60 fixed plan) but no per-second limit (10 of 15). We found no 429 or backoff guidance for Vector. Upserts by ID overwrite, so a retried write is safe (5 of 15). 99.9% uptime SLA on pay as you go and fixed, 99.99% on Pro (10). 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.

## 5. Maintenance & community, 46 out of 100, up to 4.7 more on the total

Why it scored 46: The Upstash MCP server shipped v0.3.0 on 24 August 2026 (20), with v0.2.4 to v0.3.0 since 6 July (20). Those releases didn't touch Vector, though. The npm package has no Vector tools at all, vector-js last shipped 1.2.3 on 9 March 2026, vector-py 0.8.0 on 27 February 2025, and the Vector changelog's last entry is August 2025 (less 15). Public changelog and Discord support, with no reply data for the SDK repos (5). TypeScript client current, Python client 19 months old (8). CI on vector-js and the MCP server, which publishes to the official registry on release (8).

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.

## 6. Agent ergonomics, 75 out of 100, up to 4.1 more on the total

Why it scored 75: The hosted MCP server lists 55 tools across Redis, QStash, Workflow, Vector, Search, Box and Blob, of which 17 are for Vector and Search (5). A read-only key or grant switches off every write tool (5 back). `topK`, metadata filters, `includeMetadata`, `includeVectors` and `includeData`, namespaces, cursor-based range and resumable queries (20). Error responses aren't documented as a list (10). Upserts by ID are safe to repeat, and each MCP tool is labelled read, write or destructive (20). With hosted embedding, `upsert-data` and `query-data` take plain text. Official clients for TypeScript, Python and PHP (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. Transparency & trust, 82 out of 100, up to 1.6 more on the total

Made of editorial 64, provenance 100.

Why it scored 82: Closed service with published terms. The SDKs and MCP server are MIT (18). DPA, subprocessor list and security-measures PDFs agree with the privacy PDF (25). No deprecation policy, and BGE embedding models are closed to new indexes with no date given (3). Subprocessors and regions listed, and the local MCP server's diagnostic telemetry is documented with `--disable-telemetry` (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.

## 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 server's Vector tool schemas, which aren't published
- When the hosted MCP server gained its Vector and Search tools, since no changelog entry covers it
- Whether Vector returns 429 or a plain error when the daily cap is hit

## Weaknesses

- The Vector changelog's last entry is August 2025 and the Python client last shipped in February 2025
- Vector tools exist only on the hosted MCP server, mixed into 55 tools, and not in the open-source stdio package
- No OpenAPI file and no API versioning
- SOC 2 covers Upstash Redis on Pro and Enterprise plans, not Vector
- No full-text search beyond sparse BM25 vectors; no facets, sorting or typo tolerance

## 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.

- Connect to mcp.upstash.com/mcp for Vector tools. `npx @upstash/mcp-server` has none
- Create the index with an embedding model if the agent works in text; then `query-data` and `upsert-data` take strings
- Give retrieval-only agents the index's read-only token, not the full one
- Filters use a SQL-like string (`genre = 'fiction' AND year > 2020`) in the `filter` field
- Ignore the FAQ line saying hybrid isn't supported. Pass `queryMode` HYBRID, DENSE or SPARSE on hybrid indexes

## What the review panel asked for

- dated deprecation notices
- a current Vector changelog
- Price the hosted embedding
- Document spend limits

## 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.