Mumble

# Fix list: Mumble

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/mumble, the October 2026 research run, assessed 14 October 2026. Grade F, 7.5 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 Mumble: 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, 5 out of 100, up to 19 more on the total

Why it scored 5: Graded on no interface. The docs describe a local MCP server, `mumble-mcp`, but no download, package or repository for it or for the app was found on 14 October 2026, so an outside agent has nothing to call. Read with the local-software lines, since Mumble is a Mac app with no hosted service. No official package or public download, only an early-access waitlist, though macOS 13 or later is stated (0 of 20). No public CI or test suite (0 of 25). No public issue tracker, so nothing to count. Judgement call (5 of 25). No version number, release notes or changelog found (0 of 15). Labelled beta, with no version at or past 1.0 (0 of 15). The hosted lines give the same total, 5 for no readable history and 0 on the rest.

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, 0 out of 100, up to 17.5 more on the total

Why it scored 0: Graded on the interface an outside agent can use. With none that can be obtained, the credential, least-privilege, injection and audit lines score nothing (0 of 30, 0 of 20, 0 of 15, 0 of 15). For the record, the docs say `mumble-mcp` needs no key and no network of its own and treats its start path as a boundary that `..` and symlinks can't leave, and the transcripts it would return are untrusted speech with no injection guidance found. The security programme line doesn't need an interface and is scored. security.txt returns 404 on www.tinyspace.ai, and no disclosure policy, bug bounty or certification was found (0 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. Agent ergonomics, 0 out of 100, up to 16.3 more on the total

Why it scored 0: Graded on the interface an outside agent can use. Mumble has none that can be obtained today, so context cost, pagination, errors, retries and defaults score nothing (0 on each line). For the record, the docs list 15 tools in capture mode and two more with `--refine`, prefer the batch `add_nodes`, return errors as values such as `{"ok": false, "error": "no node bata"}` and have no undo tool, recommending git instead.

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. Schema & documentation, 10 out of 100, up to 14.6 more on the total

Why it scored 10: Graded on the interface an outside agent can use, which is none, the reading used for every listing without one. The docs page describes 17 MCP tools in two modes with one-line purposes, a worked example and errors returned as values, but the binary isn't publicly available and its tool definitions can't be read, so the contract, description, typed-input and example lines score nothing (0 of 25, 0 of 20, 0 of 15, 0 of 15). tinyspace.ai/llms.txt describes the product, privacy boundaries, availability and pricing in Markdown (10). No version or changelog for the tools (0 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.

## 5. Payments & pricing, 30 out of 100, up to 8.8 more on the total

Why it scored 30: Read with the plan lines, since the app can't be downloaded and the self-hosted reading assumes a package anyone can run. No x402, MPP or L402 (0). The pricing section is public and says $0 during the beta and $0 for local use after it, while the Private (one-off) and Team (monthly) plans are marked coming soon with no price (10). Free during the beta with no card mentioned (20). Access is a waitlist that a person joins by email and confirms by link, with no programmatic route (0). The live map runs on the person's own Claude Code or Codex plan, which is that plan's cost.

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.

## 6. Maintenance & community, 0 out of 100, up to 8.8 more on the total

Why it scored 0: Read for a closed product. No release, version number, release note or dated changelog entry was found, and /mumble/changelog returns 404 (0 of 30, 0 of 20). No support address, issue tracker or community channel was found, only the waitlist form (0 of 15). No entry in the official MCP registry (a search for mumble returned none) and no SDK (0). No package or CI to judge, and `mumble-mcp` and `@mumble/core` are not on npm (0).

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.

## 7. Transparency & trust, 13 out of 100, up to 7.6 more on the total

Made of editorial 10, provenance 16.

Why it scored 13: The editorial half. Closed source, and no terms of use or licence was found, with /terms and /mumble/terms returning 404 (0 of 30). The overview, FAQ, llms.txt and the "What leaves your Mac" page agree that audio is transcribed on the Mac and never uploaded, that the audio, transcript, map and vocabulary files stay in ~/Mumble, and that transcript text and map context go to the person's Claude Code or Codex provider only when AI functions are on. No privacy policy, named controller, retention period or deletion route was found, and nothing says how waitlist emails are handled (10 of 30). No deprecation policy or dated notices (0 of 20). No statement on analytics, crash reports or update checks from the app and no opt-out for them. Transcription-only mode is credited above (0 of 20). The provenance half is computed from the provenance block, which has no legal entity, terms, changelog or security.txt.

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):

- Legal entity named: not found (0 of 20)
- Domain age: tinyspace.ai, registered 2026-07-02 (under a year) (0 of 15)
- Privacy policy: read, states 0 of the 8 things a reader expects (4 of 10)
- Status page: not found (0 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 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 people admitted from the waitlist receive `mumble-mcp` with the app. The docs call it a "developer setup today" and give only `/path/to/mumble-mcp`, and no public download, package, repository or registry entry was found on 14 October 2026. If it ships with the app, the schema, ergonomics and security lines could be read for a stdio MCP server, though it still couldn't be obtained without a person joining the waitlist.
- Whether the app needs an account, sends analytics or crash reports, or checks for updates. Nothing on the site says.
- Who the legal entity behind tiny space is. RDAP shows a GB registrant with other data redacted, and the site names none.
- The docs' "developer setup today" for a binary with no public route was considered as a misleading claim and not deducted, since the page labels it a developer setup and the product as a waitlist beta.

## Weaknesses

- No API or MCP server an outside agent can call. The documented `mumble-mcp` binary has no public download, package or repository (checked 14 October 2026)
- Access is an early-access waitlist with email confirmation, and no download of the app was found
- No terms of use, privacy policy, legal entity, changelog or security.txt was found on tinyspace.ai
- Transcript text and map context go to the person's Claude Code or Codex provider whenever AI functions are on
- English only, with no speaker labels and no shared editing, and Mac only (macOS 13 or later)

## 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.

- Don't plan to call Mumble from another agent. Neither the app nor `mumble-mcp` can be obtained publicly, so hand the task to a person with early access
- Don't configure `/path/to/mumble-mcp` from the docs. It is a placeholder, and no binary is published
- If the person already has Mumble, read ~/Mumble/<capture>/ directly, `capture.json` for the transcript and `diagram.canvas` for the map
- Keep canvas keys in schema order and never rename a node id, or the edges that name it are orphaned and the canvas is refused. Use git as the undo
- Don't write to a capture that is being recorded. The app doesn't merge the two versions

## 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.