Docusign

# Fix list: Docusign

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/docusign, the October 2026 research run, assessed 7 October 2026. Grade B, 62.5 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 Docusign: 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: Read with the hosted rubric. No x402, MPP or L402 (0). Plan prices are public without login, $50, $300 and $480 a month billed annually with a starting envelope allowance, and no per-envelope or overage price is shown (10). The developer account is free with no time limit and most capabilities enabled, on a demo environment where envelopes aren't legally binding. We didn't run the sign-up form to confirm no card is asked for (20). A person signs up in a browser, creates the integration key and grants consent, and production adds a Go-Live review (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. Reliability, 57 out of 100, up to 8.6 more on the total

Why it scored 57: Read with the hosted lines and scored on the eSignature REST API, with the MCP server noted. health.docusign.com is Docusign's own status centre with incidents by site and a history (20). Its feed lists 34 incidents since 9 July 2026. Docusign's post-mortems call two of them service disruptions, document viewing and signing failures on NA4 on 30 July (18:52 to 20:51 UTC) and failed or delayed email notifications, signing requests among them, from 25 to 30 September. Send and sign also had elevated latency and errors across NA1 to NA4 on 9 September (00:42 to 03:12 UTC). That is several majors (0). Limits have numbers, 3,000 requests an hour per account and 500 per 30 seconds in production (15). The docs describe `X-RateLimit-Reset` and burst headers, tell apps to stop until the reset, and return 429 for limit errors. `transactionId` lets a sender check whether an envelope was created after a lost response, valid for seven days, but there's no general idempotency key (12 of 15). No SLA was found in the Master Services Agreement or the terms index. The service schedules weren't readable (0). The eSignature API v2.1 is generally available, and the docs give 30 September 2026 as the MCP server's general availability date (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.

## 3. Security & auth, 64 out of 100, up to 6.3 more on the total

Why it scored 64: OAuth 2.0 with scopes, authorisation code grant with PKCE or a secret, and JWT Grant with an RSA key pair and one-hour tokens. Consent can be revoked. The `signature` scope covers most of the eSignature API, and a JWT integration can impersonate any consenting user (26 of 30). There's no read-only eSignature scope. The MCP limitations page says admins can only switch MCP access on or off, with no control over read and write tools. Envelopes can be created as drafts before sending, and production MCP keys need Docusign's approval (8 of 20). Envelope and agreement content reaches the model, and no prompt-injection guidance was found in the MCP docs. The AI attachment asks customers to review AI output (3 of 15). Each envelope has an audit_events endpoint, Connect has failure logs, a Monitor API exists, and MCP calls land in the eSignature API audit logs (15). ISO 27001:2022, SOC 1 and SOC 2 Type 2, PCI DSS, C5 and IRAP on the compliance page. security.txt returns 404, and no bug bounty or vulnerability disclosure policy was found on the trust pages. We didn't search NVD (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.

## 4. Schema & documentation, 70 out of 100, up to 4.9 more on the total

Why it scored 70: Spec files for eSignature v2.1 (Swagger 2.0, 213 paths, 414 operations) and ten other APIs are public under MIT in docusign/OpenAPI-Specifications. The MCP tool schemas sit behind sign-in, so the API carries this line (25). developers.docusign.com/llms.txt and www.docusign.com/llms.txt return 404, and no Markdown copy of the docs was found (0). Operation descriptions are long and say what a call is for (the median is 179 characters and Envelopes:create explains drafts, templates and required fields), though 43 of 414 are under 40 characters (15 of 20). The eSignature spec has no `enum` and no `required` list on any schema, and no security definitions. Valid values are listed in prose (6 of 15). How-to guides and code in six languages, an error-code page and a troubleshooting page exist, but the spec documents only a 400 with `errorCode` and `message` for most operations (12 of 15). The API is versioned (v2.1, builds such as 26.2.01.01), the SDK changelogs are dated, and the developer changelog mixes product changes with blog posts (12 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. Agent ergonomics, 70 out of 100, up to 4.9 more on the total

Why it scored 70: Envelope listing takes `count` (up to 1,000) and an `include` parameter to add recipients, folders and other detail, but there's no field selection and envelope objects are large. The MCP server lists 42 tools with no toolsets or read-only subset (15 of 25). `count`, `start_position`, `nextUri`, and filters by date, status, folder, user and search text (20). Errors carry an upper-case `errorCode` and a message, with a published code list and troubleshooting page. The docs say some errors return HTML or an empty body (15 of 20). `transactionId` guards envelope creation for seven days. There's no general idempotency key, and we couldn't see whether the MCP tools carry readOnlyHint or destructiveHint (8 of 20). Official SDKs in C#, Java, Node.js, PHP, Python and Ruby. A first call still needs the account ID and base URI from /oauth/userinfo, and an envelope definition is verbose (12 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.

## 6. Transparency & trust, 72 out of 100, up to 2.5 more on the total

Made of editorial 59, provenance 85.

Why it scored 72: The service is closed under a Master Services Agreement dated 14 November 2022. The spec files and SDKs are MIT (15). The data protection attachment of 4 September 2024 promises prompt deletion after termination without a number of days, and the privacy notice gives no retention periods. The privacy notice says systems are designed to avoid training models on customers' personal information without consent, and the AI attachment of 8 July 2026 takes that consent for AI improvement data with an opt-out toggle (20 of 30). No written deprecation policy was found. Changes are announced in dated posts, such as the 23 June 2026 notice of status-code changes rolling out to production in July (6 of 20). The sub-processor list was updated on 18 September 2026 and has an RSS feed, the attachment gives 30 days' notice of new sub-processors, and the privacy notice names Australia, the United States, the European Union and Canada as storage locations (18 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 4 of the 7 things a reader expects, and has 1 clause that costs points (5.4 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 81 out of 100, up to 1.7 more on the total

Why it scored 81: The MCP server's general availability on 30 September 2026 is the latest dated change found (30). Dated entries in the last 90 days include the Node SDK 10.0.0 on 16 July, the Slackbot MCP connector guide on 23 July, a Data IO update on 19 August and the MCP release, alongside nine blog posts (20). Closed service with a public changelog, a support centre and a developer support page. We didn't test a support channel (10 of 15). Official SDKs in six languages. The Node SDK tracks eSignature v2.1-26.2.01.01, while the stable Python package is 6.1.0 from 13 March 2026 with only release candidates since (15). The Node SDK repository has integration tests and a dated changelog, no CI workflow is visible in it, and the spec repository's last commit is 3 July 2026 (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.

## 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 service schedules linked from the terms index, so whether a paid tier has an SLA with a figure is not established
- unchecked: the MCP server's live tool definitions and annotations, which need a signed-in session (an unauthenticated initialise call returned 403)
- unchecked: whether the developer account sign-up asks for a card
- unchecked: NVD and any vulnerability disclosure or bug bounty page outside docusign.com/trust
- unchecked: per-envelope overage prices and what happens when a plan's envelope allowance runs out
- The APIs marketing page says the MCP server is in open beta while the developer docs give 30 September 2026 as its general availability date
- The JWT Grant page gives a one-hour access token while the MCP page says JWT tokens expire after eight hours

## Weaknesses

- health.docusign.com lists 34 incidents since 9 July 2026, including a two-hour NA4 disruption on 30 July and email notification failures from 25 to 30 September
- Production access needs a paid account, an admin and a Go-Live review, and MCP integration keys need Docusign's approval since 30 September 2026
- The eSignature spec is Swagger 2.0 with no enums, no required lists and no security definitions. Valid values sit in prose
- The `signature` scope covers most of the eSignature API, and MCP admins can only switch access on or off, with no read or write tool control
- No llms.txt, no security.txt, and no bug bounty, written deprecation policy or SLA found in the pages reviewed

## 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.

- Call /oauth/userinfo once after sign-in, cache `base_uri` and the account ID, and send every eSignature call to that host under /restapi/v2.1
- Create and send an envelope in one Envelopes:create call with `status` set to `sent`. Docusign asks for five calls or fewer per envelope
- Subscribe to Connect or set `eventNotification` for status. Polling one envelope more than once every 15 minutes is flagged and can fail the Go-Live review
- Set `transactionId` on envelope creation so a retry after a lost response can find the envelope. The ID is valid for seven days
- Read `X-RateLimit-Reset` and `X-BurstLimit-Remaining`, and stop calling until the reset after a 429

## 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.