Mem0 Platform + MCP by Mem0

# Fix list: Mem0 Platform + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mem0, the October 2026 research run, assessed 1 October 2026. Grade C, 56.6 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 Mem0 Platform + 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, 30 out of 100, up to 14 more on the total

Why it scored 30: A status page exists at status.mem0.ai, but it refuses automated readers, so we couldn't see its components (10 of 20) or its incident history (5, no readable history). No rate limits for the Platform API found in the docs or the OpenAPI spec (0). The spec documents no 429 and no Retry-After, and we found no retry guidance (0). Enterprise lists "SLA support" with no published terms (5 of 10). The v3 API and the hosted MCP server are generally available (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. Security & auth, 45 out of 100, up to 9.6 more on the total

Why it scored 45: Plain revocable API keys in an `Authorization: Token` header, and browser OAuth on the MCP server with no scopes documented (20 of 30). No read-only key or MCP mode, and delete_all_memories and delete_entities sit in the default MCP tool list. Bulk deletes need at least one filter, which stops an accidental wipe of everything (3 of 20). The tool returns stored memories that came from user text, and we found no prompt-injection guidance in the docs (0). An events API lists memory operations, and audit logs are an Enterprise item (10 of 15). SECURITY.md with private advisory reporting and a 72-hour acknowledgement, no published advisories, and a SOC 2 Type I claim on the pricing page. No security.txt and no bug bounty found (12 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.

## 3. Payments & pricing, 50 out of 100, up to 6.3 more on the total

Why it scored 50: No x402, MPP or L402 (0). Plan prices and monthly quotas are public, but there's no per-call or overage price (10 of 20). Hobby is free with 10,000 adds and 1,000 retrievals a month and needs no card (20). `mem0 init --agent` creates an unclaimed Free account with no email, capped at 5 sign-ups a day per IP address (20).

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.

## 4. Agent ergonomics, 63 out of 100, up to 6 more on the total

Why it scored 63: The hosted MCP server has 11 tools (15), with no toolsets or read-only subset to add back. Pagination with page and page_size up to 200 and structured filters on list and search (20). Errors are thin, with only 400 and 404 documented and no error catalogue (8 of 20). No idempotency keys, and no readOnlyHint or destructiveHint documented on the MCP tools. Writes return an event ID that get_event_status can check, which makes a retry decision possible (5 of 20). Python and TypeScript SDKs and a short required set (user_id plus messages) on add (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.

## 5. Schema & documentation, 80 out of 100, up to 3.3 more on the total

Why it scored 80: Public OpenAPI 3.0.1 spec at docs.mem0.ai/openapi.json with 16 paths (25). llms.txt with a "Use when" line for each page (10). The hosted MCP source isn't public, and the documented tool descriptions are one line each with no when-not-to-use (12 of 20). The spec uses enums for entity types and event statuses and marks required fields, but search and list filters are open objects with AND, OR and NOT operators (10 of 15). Examples throughout, but only 400 and 404 are documented as errors, with no 401, 429 or 5xx (8 of 15). v1, v2 and v3 paths, migration guides and a dated changelog (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, 66 out of 100, up to 3 more on the total

Made of editorial 56, provenance 75.

Why it scored 66: The extraction library is Apache-2.0, but graph memory runs only on the closed Platform, which is the product listed here (20 of 30). The privacy policy (updated 22 August 2026) states that Free Plan interactions train Mem0's models and paid ones don't, keeps memories for the life of the account, and the docs say account deletion completes within minutes. Retention after closure is vague, "a limited period" (18 of 30). Migration guides for v2 to v3 and a changelog that names removals, but the April 2026 removal of the graph store view carried no stated notice and there's no written deprecation policy (10 of 20). The policy names OpenAI and Anthropic as AI providers and says data may go to the United States. The trust centre at trust.mem0.ai didn't render for us and we found no full subprocessor list (8 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):

- Domain age: mem0.ai, no registry record we could read (0 of 15)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 85 out of 100, up to 1.3 more on the total

Why it scored 85: Python mem0ai 2.2.1 and npm mem0ai 3.3.1 both shipped on 2026-09-25 (30). More than ten PyPI releases since July (20). 183 open issues and 421 open pull requests. A maintainer answered the rate-limit bug in the self-hosted server (#7260) with a fix PR, but the pull-request backlog is large (15 of 25). Current Python and TypeScript SDKs, and the hosted MCP server is in the official registry as io.github.mem0ai/mem0 (15). We didn't confirm CI status on the default branch, so package health gets 5 of 10 for current majors only.

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.

- We couldn't read status.mem0.ai or trust.mem0.ai, so incident history, components and the subprocessor list are unverified.
- The MCP docs show the key as a Bearer token while the REST API uses `Token`. We patched the connect snippet to Bearer for MCP and didn't test either.
- The hosted MCP grew from 9 to 11 tools (list_events and get_event_status). Whether the tools carry readOnlyHint or destructiveHint annotations isn't documented.
- No rate limits found for the Platform API on any plan.

## Weaknesses

- Free Plan data is used to train Mem0's models, per the privacy policy of 22 August 2026
- No published rate limits or 429 handling, and only 400 and 404 documented as errors
- Retrieval caps are low below Pro, 1,000 a month free and 5,000 on Starter
- delete_all_memories and delete_entities are in the default MCP tool list with no read-only mode
- SOC 2 Type I, not Type II, and no security.txt

## 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.

- Send the REST key as `Authorization: Token <key>`, and the MCP key as Bearer
- Don't search for a memory straight after adding it. Adds are queued, so poll get_event_status with the returned event ID
- Scope every add and search with `user_id` (or agent and run IDs) so memories don't mix between users
- Pass a filter on every bulk delete, since delete_all_memories wipes whatever the scope matches
- On Hobby and Starter, count retrievals rather than adds. The retrieval quota is a tenth the size

## What the review panel asked for

- Document missing errors
- Longer tool descriptions
- scoped read-only keys
- no training on Free

## 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.