Orkes Conductor Human tasks by Orkes

# Fix list: Orkes Conductor Human tasks

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/orkes-conductor, the October 2026 research run, assessed 1 October 2026. Grade C, 54.2 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 Orkes Conductor Human tasks: 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, 32 out of 100, up to 13.6 more on the total

Why it scored 32: No public status page found for Orkes Cloud or the Developer Edition (0), so no readable incident history (5). API rate limits aren't published. The rate-limit page covers per-task execution limits you set yourself, and the Developer Edition says its limits may change (0). No 429 or backoff guidance found, but workflow starts take an idempotency key with `FAIL`, `RETURN_EXISTING` or `FAIL_ON_RUNNING` strategies (7 of 15). The Cloud support policy sets 99.9 or 99.0 per cent by plan with service credits, and Enterprise goes up to 99.99 per cent (10). The Human task is GA (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. Payments & pricing, 20 out of 100, up to 10 more on the total

Why it scored 20: No machine payment protocol (0). Paid editions are contact sales only (0). The hosted Developer Edition is free, with no card mentioned, though the pricing page says it isn't for production (20). A person signs up in the browser and creates an application key (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.

## 3. Schema & documentation, 65 out of 100, up to 5.7 more on the total

Why it scored 65: No public OpenAPI found for the Orkes API. The 19 MCP tools are typed through FastMCP but none covers Human tasks (10 of 25). llms.txt exists but is the whole documentation in one 2.57 MB file with no index, which an agent can't use as a map, so 7 of 10. Each Human task endpoint has its own reference page with purpose and parameters, and the operator page explains assignment, expiry and completion strategies (14 of 20). Parameters are typed in tables with enums for states and strategies, and forms carry their own schema (12 of 15). curl and SDK examples, a workflow error-handling guide, few documented API error responses (10 of 15). Conductor OSS releases on GitHub and a product changelog page (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.

## 4. Security & auth, 71 out of 100, up to 5.1 more on the total

Why it scored 71: Application access keys with roles and per-resource read and execute permissions, swapped for a JWT at `/api/token`. The JWT expiry is configurable, and a negative value means it never expires (25 of 30). RBAC, application roles, assignment to named users or groups, and a `TERMINATE` strategy that fails the workflow when nobody answers (18 of 20). It returns form answers from assigned reviewers (10). Human task states and history are kept per task, and we didn't find an account audit log in the docs we read (8 of 15). SOC 2 Type II named for Enterprise on the pricing page. We found no disclosure policy and couldn't check security.txt (10 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. Agent ergonomics, 69 out of 100, up to 5 more on the total

Why it scored 69: The MCP server has 19 tools and no read-only subset (15), and the Human task search pages with `start` and `size` and filters (20). Averaged to 17 of 25. Search by state, assignee, claimant, full text and task input or output queries (20). Few documented API error responses (10 of 20). Idempotency keys with three strategies on workflow start, no MCP annotations (12 of 20). Official SDKs in Java, Python, JavaScript, Go and C#, but a first approval needs a form, a workflow, a Human task and an application key (10 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, 46 out of 100, up to 4.7 more on the total

Made of editorial 41, provenance 50.

Why it scored 46: Conductor OSS and the SDKs are Apache-2.0, but the forms, assignment policies and Human Tasks API are documented for the proprietary Orkes Conductor (20 of 30). The privacy policy was last updated on 2022-02-23, keeps data 'as long as necessary', mentions no DPA and covers the website more than the cloud product. We found no retention periods for task data (8 of 30). No deprecation policy or dated notices found (5 of 20). The privacy policy says data is stored on servers in the United States, and we found no subprocessor list (8 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):

- Domain age: orkes.io, no registry record we could read (0 of 15)
- Endpoint on the vendor's domain: developer.orkescloud.com is not on orkes.io (0 of 15)
- Status page: not found (0 of 10)
- security.txt: could not be fetched (0 of 10)

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

Why it scored 81: Conductor OSS v3.32.4 on 2026-09-10, with a v3.33.0 release candidate on 2026-09-11 (30). v3.32.0 to v3.32.4 between 11 August and 10 September (20). The MCP server has 1 open issue but no commit since 2026-01-08, and the docs repository's last commit is 2026-07-06 (12 of 25). Current official SDKs in five languages (15). The MCP server's server.json still says 0.1.7 while PyPI has 0.1.9 from 2026-02-02 (4 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.

- Whether Orkes runs a public status page we didn't find.
- Data retention for Human task data and a subprocessor list, which we couldn't find.
- Whether the Developer Edition asks for a card at sign-up.
- Whether the Human Tasks API is published as an OpenAPI document on the Developer Edition server.

## Weaknesses

- No built-in Slack or email prompt, so alerts need a trigger policy and a second workflow
- Several objects to set up (form, task, workflow, application key) before the first approval
- Paid editions are contact sales only, and the free Developer Edition isn't for production
- No public status page or published API rate limits
- The MCP server has 19 tools but none for Human tasks, and no commit since January 2026

## 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.

- Give every assignment you want to escalate from a non-zero `slaMinutes`, since 0 never expires and nothing can follow it
- Pick `TERMINATE` for risky actions so an unanswered approval fails the workflow instead of staying open to anyone
- Start approval workflows with an `idempotencyKey` and `RETURN_EXISTING`, so a retried start doesn't ask twice
- Refresh the JWT from `/api/token` with a set expiry instead of requesting one that never expires
- Fetch single doc pages, not llms.txt, which is the whole documentation in 2.57 MB

## What the review panel asked for

- dated notices for Orkes Cloud changes
- maximum token lifetime
- account audit log

## 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.