UiPath

# Fix list: UiPath

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/uipath, the October 2026 research run, assessed 14 October 2026. Grade B, 68.2 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 UiPath: 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, 25 out of 100, up to 9.4 more on the total

Why it scored 25: No x402, MPP or L402 (0). Basic is published at $25 a month and the Community plan is free, while Standard, Enterprise and extra robots are priced by sales (10). The Community plan is free with no time limit by direct sign-up. No page we read says whether a card is needed (15 of 20). A person signs up in a browser and creates the PAT or external application (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, 62 out of 100, up to 7.6 more on the total

Why it scored 62: Graded on the hosted Orchestrator API and MCP servers. Public Atlassian Statuspage at status.uipath.com with component and incident history (20). The 90-day feed has several major incidents, including robot job failures in Europe from 10:00 to 17:23 UTC on 10 September 2026, solution deployments failing from 15:18 to 20:45 UTC on 11 September, Cloud Robot jobs stuck pending in the US for 94 minutes on 2 October and an expired certificate taking down several ML services across regions on 13 October (0). Per-endpoint rate limits with numbers (15). 429 responses carry Retry-After and X-RateLimit-Remaining, with retry guidance. No idempotency key for StartJobs was found (12 of 15). The Automation Cloud page states a 99.9 per cent uptime SLA, but the SLA document with credits was not found publicly (5 of 10). The Orchestrator API and the UiPath MCP server type are generally available, with Platform and Swagger servers in preview (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. Agent ergonomics, 73 out of 100, up to 4.4 more on the total

Why it scored 73: An MCP server holds only the tools a person chose for one folder, and the API takes $select, $top and $expand (22 of 25). $filter, $orderby, $top and $count, with export endpoints for bulk reads (20). HTTP codes plus numbered custom errors such as 1002 ItemNotFound and 4502 for the daily export limit (16 of 20). No idempotency key for starting jobs, and no readOnlyHint or destructiveHint described for MCP tools (6 of 20). Every folder call needs a folder header and StartJobs needs a release key. A Python SDK and CLI exist; a second official SDK language was not confirmed (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.

## 4. Schema & documentation, 79 out of 100, up to 3.4 more on the total

Why it scored 79: The Orchestrator API is described in OpenAPI 3.0, served per tenant at the Orchestrator URL plus /swagger/index.html rather than as a public file (20 of 25). llms.txt at docs.uipath.com and a Markdown copy of every page by adding .md (10). API pages state scope and limits clearly, such as which fields GET /odata/Jobs omits. MCP tool descriptions come from the description of each published artifact, so quality depends on the owner (12 of 20). OData types and an enumerated types reference. MCP tool inputs follow each process's arguments (11 of 15). Request examples on each API page and a table of custom error codes (14 of 15). Monthly dated release notes. UiPath says the Swagger version number should not be relied on (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. Security & auth, 82 out of 100, up to 3.2 more on the total

Why it scored 82: MCP OAuth with resource indicators binding a token to one server, external apps with OAuth scopes such as OR.Jobs, and PATs with chosen scopes and an expiry of up to one year (30). Folder roles limit what a token can do, and automatic client registration is off until an administrator enables it in the AI Trust Layer. No approval step before destructive API calls was found (14 of 20). Tools return job outputs from robots that read third-party applications, and no prompt-injection guidance was found in the MCP pages (5 of 15). Orchestrator audit logs and a preview API audit view (13 of 15). security.txt valid until 2028 with a HackerOne programme, and SOC 2 Type 2, ISO/IEC 27001 and FedRAMP Moderate claimed on the Automation Cloud page (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.

## 6. Transparency & trust, 76 out of 100, up to 2.1 more on the total

Made of editorial 57, provenance 94.

Why it scored 76: Closed service with published terms pages, though the MSSA that governs paid use is signed per customer and not public (15). The privacy policy covers cloud product data, says it may be used for product improvement and machine learning under the customer contract and warns against personal data on Community. The legal page dates it 30 November 2018, and the DPA download was not read (15 of 30). A product lifecycle policy and a dated deprecation timeline for the Swagger 2.0 specification (17 of 20). Hosting regions are named, such as Europe for Basic. No subprocessor list was read (10 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: published, but our reader couldn't read it (7 of 10)
- Privacy policy: read, states 7 of the 8 things a reader expects, and has 1 clause that costs points (7.3 of 10)

## 7. Maintenance & community, 80 out of 100, up to 1.8 more on the total

Why it scored 80: Orchestrator release notes dated 12 October 2026 (30). Entries on 5, 7 and 12 October and nine dates in September (20). Public dated release notes, a forum and a support portal. Support was not tested (12 of 15). The Python SDK `uipath` 2.14.37 was published on 12 October 2026. UiPath's MCP servers are not in the official MCP registry, where a search returned only a third-party server (12 of 15). The SDK's CI was not checked (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 text of the Automation Cloud SLA and its service credits; only the 99.9 per cent claim on the product page and status page was read
- unchecked: the Data Processing Agreement and a subprocessor list
- unchecked: whether Community sign-up asks for a card
- unchecked: the GitHub repository for the `uipath` Python SDK, which the GitHub API did not return, so CI and issue response were not assessed

## Weaknesses

- Standard and Enterprise prices are not published, and robot licences are sold through sales
- Status history shows robot job failures in Europe for over seven hours on 10 September 2026 and a certificate expiry hitting several services on 13 October 2026
- No idempotency key was found for StartJobs, and the MCP pages say nothing about readOnlyHint or destructiveHint
- The OpenAPI document is served per tenant behind the Orchestrator URL, not as a public file
- Tools backed by Integration Service activities time out with PATs or client credentials and need user-context sign-in

## 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 `Authorization: Bearer <token>` on every MCP request. The `mcp-session-id` header carries no authentication
- Give a PAT both `OR.Execution` (list tools) and `OR.Jobs` (run tools), or tool listing or calls fail
- Use client credentials from an external application for unattended runs. It is the only method that needs no person at sign-in
- Add a folder header such as `X-UIPATH-FolderKey` to Orchestrator API calls on folder resources
- Read job outputs with `GET /odata/Jobs({key})`. `GET /odata/Jobs` returns InputArguments and OutputArguments as null and is limited to 100 calls a minute per tenant

## 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.