Nutshell

# Fix list: Nutshell

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/nutshell, the October 2026 research run, assessed 8 October 2026. Grade D, 49.6 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 Nutshell: 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, 42 out of 100, up to 11.6 more on the total

Why it scored 42: Read with the hosted lines. Status page at status.nutshell.com with 12 components and incident history back to 2017 (20). One incident in the last 90 days, a database cluster fault on 28 September 2026 that left the application unavailable for some customers, marked major, mitigated after 43 minutes and closed after about seven hours. Scored between the minor and one-major bands because the outage was partial and under an hour to mitigation (12 of 30). No rate limit numbers for REST, and the JSON-RPC docs say only that limits vary (0). No 429, Retry-After, backoff or idempotency guidance found (0). No SLA found, and the terms supply the service as is (0). The REST API is GA and the MCP server carries no beta 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. Agent ergonomics, 44 out of 100, up to 9.1 more on the total

Why it scored 44: The MCP server is read-only with about 17 groups of search and list tools per the help centre. The tool definitions sit behind OAuth and were not read. REST lists take `page[limit]` with no field selection (13 of 25). `page[page]`, `sort`, `q` and `filter[...]` keys, with field lists from `/accounts/list/fields`, and no stated maximum page size (16 of 20). Errors are barely documented and an unauthenticated REST call returned an HTML page (4 of 20). No idempotency keys. Deletes can be undone within 30 days and the MCP server cannot write (6 of 20). Few required fields, and no official REST SDK. The only client is a PHP JSON-RPC sample last pushed in 2023 (5 of 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.

## 3. Payments & pricing, 35 out of 100, up to 8.1 more on the total

Why it scored 35: No x402, MPP or L402 (0). Plan prices are public, $13 to $79 a seat a month billed yearly, with nothing per call (10). 14-day trial with no card, and the API is included (20). A person signs up in a browser and an admin creates the key. The docs publish sandbox credentials for a shared demo account, which lets an agent test calls with no signup but not reach its own data (5 of 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. Security & auth, 58 out of 100, up to 7.4 more on the total

Why it scored 58: The MCP server uses OAuth authorisation code with PKCE, dynamic client registration, a revocation endpoint and `read` and `write` scopes. REST uses unscoped API keys sent as an HTTP Basic password with a user's email, and a key can be allowed to impersonate any user (24 of 30). The MCP server is read-only and OAuth has a `read` scope, but no read-only API key type was found (12 of 20). Records hold emails, notes and form submissions, with no injection guidance found (3 of 15). An audit log on Business and Enterprise covers logins, bulk edits and exports, and API changes are logged under the key name or the impersonated user (8 of 15). SOC 2 Type 1 from December 2025, a CASA assessment, a disclosure address, no bounty and no security.txt (11 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.

## 5. Schema & documentation, 57 out of 100, up to 7 more on the total

Why it scored 57: Each of the 114 REST reference pages carries the OpenAPI 3.0 definition of its operation, but whole-spec download is disabled (18 of 25). `llms.txt` and Markdown for every docs page (10). 78 of 114 operations have a description, and the newer ones state scope and visibility rules (11 of 20). Enums on sort fields and patterns on IDs, but filters and custom fields are open objects and `page[limit]` is typed as a string on some operations (9 of 15). Schema examples are present, and error responses appear on 9 operations with no error body schema (5 of 15). The spec is version 2.0.0 and pages carry update stamps, with no dated API changelog (4 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.

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

Why it scored 46: The newest REST reference page was updated on 21 August 2026, 48 days before the check (20 of 30). Five reference pages changed between 5 and 21 August 2026, for quote and invoice status, to-dos and dashboard fields. These are page stamps, not changelog entries (20). No API changelog, and the product updates page is undated. Live support on every plan per the pricing page, untested (6 of 25). No current official SDK, and the MCP registry was not checked (0). No packages to judge (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, 71 out of 100, up to 2.5 more on the total

Made of editorial 56, provenance 86.

Why it scored 71: Closed service with clear terms naming Nutshell, Inc. (15). The terms, privacy policy and DPA, all dated 2 June 2026, agree on a 30-day export window after termination, deletion from production afterwards, backup copies that may persist indefinitely and no training of generalised AI models on customer data (24 of 30). The JSON-RPC API is marked legacy with a promise of indefinite support, but there is no deprecation policy and the terms allow functions to be removed without prior notice (5 of 20). The subprocessor list at trust.nutshell.com could not be read. The policy names AWS and regions, and the DPA promises 10 days' notice of a new subprocessor (12 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):

- Terms of service: read, states 7 of the 7 things a reader expects, and has 2 clauses that cost points (6 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.

- unchecked: the subprocessor list and SOC 2 detail at trust.nutshell.com, a page drawn by script
- unchecked: the MCP server's tool definitions, count and annotations, which need an OAuth login
- unchecked: the API key permission types, which the help article shows only in an image
- unchecked: whether the MCP server is in the official MCP registry
- unchecked: REST behaviour on 429 and the JSON error body, since no authenticated call was made
- No REST rate limit numbers, SLA, API changelog or deprecation policy were found in the reviewed documentation
- The lead was right about API access on every plan. It did not mention the hosted MCP server, the legacy JSON-RPC API or that the vendor entity is Nutshell, Inc. in Harrisburg, with WebFX named in the site footer

## Weaknesses

- No rate limit numbers, 429 handling or Retry-After found for the REST API
- Error responses are documented on 9 of 114 operations, with no error body schema, and an unauthenticated REST call returns an HTML page
- API keys have no scopes or read-only type, and the key travels as an HTTP Basic password with a user's email
- No dated API changelog, no official SDK for REST and whole-spec OpenAPI download is disabled
- A database fault on 28 September 2026 left the application unavailable for some customers, marked major on the status page

## 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.

- Authenticate REST calls with HTTP Basic, a Nutshell user's email as username and the API key as password, against `https://app.nutshell.com/rest`
- Use IDs in the `<number>-<type>` form, such as `3-accounts`, and send updates as JSON Patch with `content-type: application/json-patch+json`
- Page lists with `page[limit]` and `page[page]`, which is 0-based, and call `/accounts/list/fields` or the lead and contact equivalents for valid filter keys
- Use the REST API for writes. The MCP server only reads, and OAuth tokens need the `read` scope
- Filter webhook events yourself, because every webhook receives all events and subscriptions are created in Setup, not through the API

## 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.