Paragon ActionKit + MCP by Paragon

# Fix list: Paragon ActionKit + MCP

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/paragon, the October 2026 research run, assessed 1 October 2026. Grade D, 47.8 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 Paragon ActionKit + 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. Payments & pricing, 0 out of 100, up to 12.5 more on the total

Why it scored 0: No x402, MPP or L402 (0). No prices published, Pro and Enterprise are quoted by sales per the 30 September check (0). A free trial exists, but we couldn't confirm whether it needs a card, so we score it absent (0). A person signs up and configures signing keys in the dashboard (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. Agent ergonomics, 41 out of 100, up to 9.6 more on the total

Why it scored 41: Thousands of tools across 130+ integrations, so 5, plus 10 back because `categories` filters the API list and the MCP server takes `LIMIT_TO_INTEGRATIONS` and `LIMIT_TO_TOOLS` (15). Category and format filters on the tool list, no pagination or response sizing documented (10). Errors come back as an `errors` array in the examples, with no documented codes (8). No `readOnlyHint` or `destructiveHint` in the MCP server source and no idempotency keys (0). One JavaScript SDK, and a run needs only a tool name and parameters (8).

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. Reliability, 60 out of 100, up to 8 more on the total

Why it scored 60: Atlassian Statuspage at status.useparagon.com with ten components, including ActionKit, Connect API, Workflows and Managed Sync (20). One incident in the last 90 days, degraded Workflows and Connect API performance on the EU cloud for about three hours on 29 July 2026 (20). Workflow concurrency is published (5 step executions on trial, 20 on Pro, 50 on Enterprise cloud), but no request rate for ActionKit (5). Workflows retry upstream 429s, per the 30 September check, and ActionKit's own 429 behaviour isn't documented (5). No SLA found (0). ActionKit is generally available. Only the Triggers API is marked beta (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. Security & auth, 65 out of 100, up to 6.1 more on the total

Why it scored 65: Every call carries an RS256 JWT your server signs per end user, so each call is bound to one user's connected accounts and you set the expiry. The project signing key can mint a token for any user (25). Tool lists can be limited by integration or name, with no read-only mode and no confirmation before writes (10). Actions return third-party content and we found no prompt-injection guidance (3). Event Logs record integration, sync and agent actions with timestamp, trace ID, user ID, credential ID and status (13). SOC 2 Type II, HIPAA, penetration tests twice a year and a trust centre at security.useparagon.com. ISO 27001 is listed as coming soon. No security.txt or disclosure policy found (14).

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. Maintenance & community, 48 out of 100, up to 4.6 more on the total

Why it scored 48: The `@useparagon/connect` SDK was last released on 23 September 2026 per the 30 September check, and the MCP repository took a fix on 30 September (30). No tagged releases or dated changelog entries found in the last 90 days (0). The changelog has been quiet since April 2026, though MCP pull requests were merged in July and September (6). One official SDK, in JavaScript (8). The MCP server has 29 tests but its only workflow publishes the Docker image, so nothing runs them in CI (4).

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. Schema & documentation, 73 out of 100, up to 4.4 more on the total

Why it scored 73: OpenAPI 3.0 file for ActionKit (version 1.0.0), and the tools endpoint returns JSON Schema for every action (25). llms.txt per the 30 September check (10). Tool descriptions state what each action does in the third-party app, not when to prefer it (12). Inputs typed per action with required fields (12). Code examples in the docs, but the OpenAPI file defines no error responses or schemas (6). The product-updates changelog is monthly, but its newest entry is April 2026 (8).

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.

## 7. Transparency & trust, 65 out of 100, up to 3.1 more on the total

Made of editorial 43, provenance 86.

Why it scored 65: Closed platform under published terms. The MCP server's README says MIT, its package.json says ISC, and there's no `LICENSE` file (13). Credentials sit in a separate vault under per-credential keys, and cloud tenants choose US or EU. No retention periods or subprocessor list on the security page (15). No deprecation policy or dated notices found, apart from the MCP README calling its SSE endpoints deprecated (5). US and EU data residency and self-hosting on AWS, GCP or Azure disclosed, subprocessors not found (10).

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: useparagon.com, registered 2019-08-30 (7 years) (11 of 15)
- security.txt: not found (0 of 10)

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- As of the 30 September 2026 commit, the self-hosted MCP server defaults `NODE_ENV` to development, and in development `/mcp` and `/sse` sign a user token for whatever ID arrives in `?user=`. The Dockerfile and the image its workflow publishes don't set `NODE_ENV`, so a server started from that image without the variable lets anyone who can reach it impersonate any end user. The README documents the behaviour and the compose file sets production, so the deduction is modest (https://github.com/useparagon/paragon-mcp/blob/main/src/index.ts, https://github.com/useparagon/paragon-mcp/blob/main/src/utils.ts).

## 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: whether the free trial needs a card
- Which licence the MCP server is under, MIT per the README or ISC per package.json
- Whether ActionKit has a request rate limit and what a throttled call returns
- How many tagged SDK releases shipped in the last 90 days

## Weaknesses

- No published prices and no self-serve paid plan
- The self-hosted MCP server runs in development mode unless `NODE_ENV=production` is set, and then trusts `?user=`
- No error schemas in the OpenAPI file, no ActionKit rate limit and no tool annotations
- Changelog's newest entry is April 2026
- No security.txt, and the MCP server's licence is stated two ways with no `LICENSE` file

## 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.

- Sign a short-lived User Token per end user on your server and never let the model see the signing key
- Set `NODE_ENV=production` before exposing the MCP server anywhere but localhost
- Fetch the tool list once per session with `categories` set, and cache it
- Set `LIMIT_TO_TOOLS` on the MCP server to keep write actions out of a read-only agent
- Proxy API requests count as tasks, so prefer ActionKit tools for routine actions

## What the review panel asked for

- tagged MCP releases
- dated deprecation for SSE
- production as the default
- read-only tool mode

## 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.