Shortcut

# Fix list: Shortcut

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/shortcut, the October 2026 research run, assessed 8 October 2026. Grade C, 60.2 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 Shortcut: 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). Plan prices are public, Free $0 for up to 10 users, Team $8.50 a user a month billed yearly or $10 monthly, Business $12 or $16, Enterprise by quote. API calls aren't priced (10). The Free plan lists API and webhook access and the MCP server, and the 14-day trial needs no card (20). A person signs up in a browser and creates a token or approves the OAuth grant (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, 54 out of 100, up to 8.1 more on the total

Why it scored 54: The hosted MCP server uses the OAuth authorisation code flow with PKCE (S256), dynamic client registration and the scopes read, write, story-write, comment-write and admin. REST tokens are created per user, can be read-only or read-write since 20 January 2026, and the v4 alpha lets an admin list and disable workspace tokens. The v3 docs still accept the token as a `token` query parameter, marked deprecated, which costs 10 (20 of 30). Read-only tokens, a read scope, narrow story and comment scopes and the view-only Observer role. No confirmation step for deletes was found (14 of 20). Stories, comments and docs written by other people reach the model, and no injection guidance was found (3 of 15). Story history is readable by API. No workspace audit log of API or MCP calls was found in the pages read (5 of 15). No security.txt (404). A disclosure policy promises acknowledgement within a week, and the security page states SOC 2 Type 2 with the report on request. No bug bounty or public advisories 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. Reliability, 64 out of 100, up to 7.2 more on the total

Why it scored 64: Graded on the hosted lines for REST API v3 and the hosted MCP server. status.shortcut.com is a Statuspage site with API and Shortcut MCP components and history back to March 2023 (20). In the 90 days to 8 October 2026 it shows three incidents, the MCP server unavailable for about 6 minutes on 14 July, tokens newly created through the MCP server unable to write for 2 hours 21 minutes on 15 July (rated major by Shortcut, fix in place after 22 minutes), and search indexing behind on 4 August. None names the API component. One vendor-rated major that affected only new MCP tokens sits between the minor and major lines (15 of 30). The docs give 200 requests a minute (15). A 429 is documented, with no Retry-After header, backoff guidance or idempotency keys found (4 of 15). The Enterprise plan lists 'Premier support SLAs', and no uptime SLA was found (0). v3 is described as the current version with backwards-compatible changes only, and the hosted MCP server carries no beta label. v4 is alpha and isn't graded (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. Agent ergonomics, 57 out of 100, up to 7 more on the total

Why it scored 57: Graded on the API. Search takes `detail=slim` to drop descriptions and comments, and list responses use slim types. Field selection (`fields`) exists only in the v4 alpha (17 of 25). Search has operators, `page_size` up to 250 and a `next` token, and epics have a paginated list, but the v4 migration note says v3 list endpoints return all results at once (14 of 20). Errors are HTTP codes with one-line meanings. A live 401 returned JSON with `message` and `tag`, which the reference doesn't catalogue (9 of 20). No idempotency keys. `external_id` fields exist on some entities. The archived MCP source (v0.25.0) sets readOnlyHint, destructiveHint and idempotentHint on its 78 tools, and the hosted server's annotations were not read (8 of 20). A story needs only a name and a workflow state. One official SDK, `@shortcut/client` for JavaScript and TypeScript (9 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.

## 5. Schema & documentation, 75 out of 100, up to 4.1 more on the total

Why it scored 75: Swagger 2.0 and OpenAPI 3.0 files for v3 are downloadable from the docs (143 operations on 85 paths), and an OpenAPI 3.0.3 file for the v4 alpha (247 operations) is at www.shortcut.com/openapi.json. The hosted MCP server's tool schemas need an account and were not read (25). www.shortcut.com/llms.txt links the spec, and help centre pages have Markdown copies. developer.shortcut.com has no llms.txt and is a single HTML page per version (7 of 10). 127 of 143 operations carry a description, mostly one line, with a few usage rules such as the `workflow_state_id` or `project_id` choice on story creation (12 of 20). Enums, maxLength, formats and required flags throughout (13 of 15). Every endpoint has a curl example and an example response, but errors are the same three lines everywhere (400 schema mismatch, 404, 422) with no documented error body (9 of 15). Versions are in the path (v2 deprecated, v3 current, v4 alpha). No API changelog was found, although the v4 page refers to one, and API changes appear in the product release notes (9 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, 73 out of 100, up to 2.4 more on the total

Why it scored 73: The newest release notes are dated 18 September 2026 and `@shortcut/client` 3.4.1 was published on 22 September 2026 (30). Release notes on 31 July, 14 August and 18 September, and four client releases in September (20). Closed service with public release notes, a community Slack and a support address. Response times were not checked (9 of 15). The official client is current, but the official MCP registry has only third-party Shortcut servers (com.mcparmory/shortcut, io.github.stayce/shortcut-mcp) (8 of 15). The client repository was last pushed on 1 October 2026 under MIT. The MCP server repository is archived at v0.25.0 (23 June 2026) with a notice that development moved to the hosted server (6 of 10).

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, 73 out of 100, up to 2.4 more on the total

Made of editorial 57, provenance 88.

Why it scored 73: Closed service under terms naming Shortcut Software Company, with an MIT client library and the archived MIT MCP server (15). The privacy policy of 11 July 2025 covers the platform and says inputs and outputs aren't used to train models by default. Retention is 'only for as long as it serves the purpose', the terms give at least 30 days of data access after termination, and the DPA (last updated 23 August 2018) is sent on request, not published (17 of 30). No deprecation policy found. v2, the Projects endpoints and the token query parameter are marked deprecated with no dates (5 of 20). The sub-processor list, updated 12 November 2025, names 33 providers with product, type and country, hosting on AWS in the US (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 5 of the 7 things a reader expects (8.3 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 hosted MCP server's tool list, schemas and annotations, because tools/list needs a Shortcut account (the endpoint answers 401 without a token). The 78 tools counted are in the archived source at v0.25.0, so `toolCount` is left empty.
- unchecked: whether the read-only and read-write tokens of 20 January 2026 work on v3 as well as v4. The v4 page describes the `sct_ro_` and `sct_rw_` prefixes and says v3 tokens don't work on v4.
- unchecked: whether the live API sends a Retry-After header on 429. None is documented.
- unchecked: the DPA (last updated 23 August 2018) and the SOC 2 report, which are sent on request only.
- unchecked: response times in the community Slack and from support.
- No API changelog was found, although the v4 page tells readers to review one. No workspace audit log was found in the help centre pages read.
- The lead said the MCP server was unconfirmed. Shortcut runs a hosted one at https://mcp.shortcut.com/mcp, and the open-source `@shortcut/mcp` repository is archived. The lead also missed the v4 alpha.
- The Enterprise plan lists 'Premier support SLAs'. Whether an Enterprise contract carries an uptime commitment was not established.

## Weaknesses

- The v3 docs still allow the API token as a `token` query parameter, marked deprecated with no removal date
- No idempotency keys, Retry-After header or backoff guidance found. The docs state only 200 requests a minute and a 429
- No API changelog or deprecation policy found. API changes appear as lines in the product release notes
- The open-source MCP server is archived at v0.25.0, and the hosted server's tool list can't be read without a Shortcut account
- One official SDK, `@shortcut/client` for JavaScript and TypeScript, and no vendor entry in the official MCP registry

## 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 v3 token in the `Shortcut-Token` header. v4 (alpha) takes `Authorization: Bearer` with `sct_ro_` or `sct_rw_` tokens, and v3 tokens don't work there.
- Create a story with `name` and `workflow_state_id`. Sending both `workflow_state_id` and `project_id`, or neither, is rejected.
- Use `GET /api/v3/search/stories` with `detail=slim`, `page_size` (1 to 250) and the `next` token. Many other v3 list endpoints return every record at once.
- Stay under 200 requests a minute and add your own backoff on 429, because no Retry-After header is documented.
- For MCP, connect to https://mcp.shortcut.com/mcp and request only the scopes needed, such as `read` or `story-write`.

## 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.