Hermes Agent vs OpenWork: Cards or Versioned Scripts?

5 min read Comparisons

Compare Hermes Agent and OpenWork on profile coordination, saved workflows, permissions and what their hosted prices cover.

Hermes Agent and OpenWork put a different artifact at the center of repeatable work: dependency-linked cards or a tested, versioned script. For a developer choosing an agent environment, the deciding question is whether the next job needs specialist handoffs or an exact saved program with validated inputs and outputs.

Both publish scheduling, reusable knowledge and local execution. Those shared capabilities leave an important distinction unresolved: coordinating agents through dependencies does not establish that the same process will execute unchanged on every run.

Hermes Agent coordinates named profiles

Hermes Agent's Kanban documentation describes an orchestrator profile that owns a project's root card, delegates child cards to specialists and judges whether the project is complete. Cards retain comments, dependencies and attempt history. People manage the board and contribute comments, while the documented assignee is a profile name.

That model suits a developer dividing a research or implementation job into pieces whose completion needs judgment. You can also stop automatically generated child tasks from being promoted until you review them. The orchestrator can add work when its completion review uncovers a gap.

Dependency chains provide ordered handoffs, but the same documentation reserves workflow-template routing for a future version. The kernel ignores those routing fields. Evaluate the published Kanban coordination on its own merits, without treating the template fields as an executable workflow definition.

OpenWork preserves a successful script

OpenWork runs the OpenCode engine through desktop and browser interfaces. Its Workflow saving API starts from a successful Code Mode run. A saved Workflow captures a script you can execute again with new inputs and live tools.

The version execution documentation describes running an exact saved version, checking input and output schemas and recording an execution receipt. This is a concrete fit when you have already worked out a useful script and want subsequent executions to follow that version.

OpenWork's surrounding work stays in chats that you can organize, archive and restore. Its public pages do not describe those chats as work records assignable to both people and agents. Schema checks apply to saved scripts, so do not assume each ordinary chat task receives an acceptance-criteria check.

Decide what completion should prove

Hermes Agent's goal documentation describes completion contracts and shell quality gates. A goal can require a test result before it finishes, and Kanban review can match acceptance criteria to evidence and request correction. This accommodates a job whose implementation changes while its completion condition remains fixed.

OpenWork requires a successful tested run before saving a script as a Workflow and rejects execution data that violates the saved schemas. Those checks answer whether the script meets its input and output contract. They do not establish a review against every requirement you might put into a chat request.

For a weekly repository report, distinguish valid output structure from a satisfactory report. A schema may establish that a result has the expected fields. Acceptance criteria might also require evidence for each conclusion. Use the relevant product's published check for the question it actually answers, and define any remaining human review before starting.

Reuse knowledge and permission decisions

Hermes Agent keeps persistent notes and preferences that you can inspect and change, according to its memory guide. Its security guide also describes an approval-history command that ranks allowlist proposals from executed, approved commands. You select which entries to apply, and destructive command classes are excluded.

OpenWork can turn a completed chat into an editable skill. Its cross-chat memory guide describes searching saved sessions within the current app and workspace context. Transcript reads return recent messages, so older portions of a long exchange may not be returned.

OpenWork's Enterprise Agent permissions can allow, request approval for or block commands, edits and websites. These controls are rolling out and require the OpenCode v2 engine. Its public pages do not describe proposals to change those permissions from approval history, so reusable skills and permission learning should be evaluated separately.

Account for execution and billing

Hermes Agent offers a free MIT agent and optional hosted execution. Hermes Cloud lists Plus $20/month, Super $100/month and Ultra $200/month, with credit top-ups. Its FAQ separately meters compute and storage and excludes inference and tools, without explaining how those charges relate to included plan hosting. Confirm that relationship for the configuration you intend to use.

OpenWork's pricing page lists a free desktop app and Free covering five users. Team is $10 per user per month, billed monthly, and Enterprise is custom. Cloud computers add $50 per member monthly and models $10 per user monthly. These are seat plans and add-ons, so compare their covered resources with the Hermes configuration before choosing a hosting path.

Hermes Agent and OpenWork, point by point

The table presents both products' answers to the same questions. Each answer includes its source and check date, including narrower capabilities and rollout conditions.

Both columns answer the same questions from each company's public pages. Yes means the company publishes the capability the question asks about, Partly means something narrower, No means it says it does not offer it, and Not published means its public pages do not describe it. Every answer links to its source and shows when we last checked it. The oldest check on this page was November 13, 2026.

Hermes Agent and OpenWork compared, with the source and check date of each answer
What you compare Hermes Agent OpenWork
Goals Hermes Agent Partly

Persistent goals keep an agent working toward an objective, with a judge and optional completion tests. Its docs describe no linked metric tracking against an ongoing business goal.

OpenWork Not published

Its public docs do not describe an agent or named owner tracking a measurable goal and reviewing progress against its target.

Leads Hermes Agent Yes

You can name an orchestrator profile to own a decomposed project's root task, assign child tasks to specialists, judge completion and add more work when needed.

OpenWork Not published

Its docs describe administrative owners and a custom Team lead permission role, but no person or agent lead responsible for reviewing and coordinating project, goal or team work.

Tasks Hermes Agent Partly

Kanban cards keep an assigned agent profile, comments, dependencies and attempt history. People can comment and manage cards, but the documented assignee is a profile name rather than a person or an agent.

OpenWork Partly

Work stays in chats you can name, group, archive and restore, and Automation runs keep history. Its docs do not describe work records assignable to both people and agents.

Recurring work Hermes Agent Yes

Cron jobs run once or recur on a schedule, and webhook events can fire the same jobs immediately. Each job can carry its own instructions, skills and delivery settings.

OpenWork Yes

Automations run once, daily, weekly or around Google Calendar meetings, on desktop or in Cloud. An incoming Slack mention or direct message also starts agent work, but that bot is private alpha for approved organizations.

Workflows Hermes Agent Partly

Kanban dependency chains coordinate roles such as scout, editor and writer. They provide ordered handoffs, but the kernel does not execute workflow templates, whose routing fields are reserved for a future version.

OpenWork Yes

A successful Code Mode script can become a reusable Workflow, and an exact saved version runs with supplied inputs and live tools. Its API records an execution receipt and validates input and output schemas.

Autonomy Hermes Agent Yes

You configure shell approval modes, permanent allowlists and deny rules. Dangerous-command requests appear in the terminal or messaging chat, where you can approve or deny them.

OpenWork Yes

Enterprise Agent permissions can allow, ask first or block commands, file edits and websites, and members answer permission requests in their sessions. This feature is rolling out and requires the OpenCode v2 engine.

Learned autonomy Hermes Agent Yes

The approvals suggest command mines executed, approved commands into ranked allowlist proposals. You choose which entries to apply, and destructive command classes are excluded.

OpenWork Not published

Its docs do not describe proposals to change action permissions based on approval history. Editable skills and administrator-set permission rules do not establish that behavior.

Memory Hermes Agent Yes

Persistent notes and user preferences carry into later sessions. You can inspect, edit or delete learned memory through the journey commands or the desktop panel.

OpenWork Yes

You can turn a completed chat into a reusable skill, edit its instructions in chat or Cloud and use it in later work. Cross-chat recall also searches saved session history rather than a separate long-term memory database.

Planning Hermes Agent Yes

From chat, Kanban can expand a rough idea into a specification or dependency-linked tasks. You can disable automatic child promotion so the generated tasks wait for manual review before execution.

OpenWork Yes

OpenWork runs the OpenCode engine and uses its configuration. OpenCode's Plan agent explores without ordinary file edits and can write plan files when asked, so planning can produce an artifact before execution.

Intake Hermes Agent Yes

Signed webhooks turn outside events into agent prompts. A route can target a named profile when profile multiplexing is enabled, and filter which events start work.

OpenWork Yes

In its private-alpha Slack bot, a mention or direct message reaches the invoking member's assistant with that member's connected-service access. Each member gets a separate session, even in the same Slack thread.

openworklabs.com/…/overview.md
Connectors Hermes Agent Yes

A curated Model Context Protocol catalog and custom local or remote servers connect external tools. HTTP webhook routes accept signed service events and can post responses back through integrations such as GitHub.

OpenWork Partly

A catalog and custom MCP addresses connect Google Workspace, Microsoft 365 and Slack. Its connector guide does not describe custom HTTP/API connector definitions or general webhook triggers, and incoming Slack bot requests remain private alpha.

Models Hermes Agent Partly

You choose the main model and can override auxiliary tasks. Provider routing chooses a backend for a selected model, rather than selecting a model from each task's needs.

OpenWork Yes

Smart Routers choose a model and thinking level for each prompt from administrator-selected AI Gateway models tagged by task. The following tool steps keep that model and thinking level until the task ends.

Cloud or your own computer Hermes Agent Yes

Run the free agent locally through its desktop or terminal interfaces, or choose the optional Hermes Cloud service for a hosted instance with a persistent workspace.

OpenWork Yes

The desktop app runs on your computer, OpenWork Web offers a cloud computer in a browser, and the platform can be self-hosted. The browser computer is a separate paid add-on.

Your own subscription Hermes Agent Yes

Hermes accepts your provider API keys and a ChatGPT account login. Its Claude account path requires Max plus purchased extra-usage credits, not the included Max allowance.

OpenWork Yes

The desktop app accepts your provider API key, ChatGPT sign-in or local model. Team plans also distribute your own provider keys through the organization.

Pricing Hermes Agent Yes

The MIT agent is free. Portal lists Plus $20/month, Super $100/month and Ultra $200/month, with credit top-ups. Cloud's FAQ separately meters compute and storage, excluding inference and tools, without explaining how that relates to included plan hosting.

OpenWork Partly

Desktop is free and Free covers five users. Team is $10 per user per month, billed monthly, and Enterprise is custom. Cloud computers add $50 per member monthly and models $10 per user monthly. These are seat plans, not usage billing.

Correctness Hermes Agent Yes

Goal completion contracts define verification criteria, and shell quality gates must pass before a goal completes. Kanban review also maps acceptance criteria to evidence and can return work for correction.

OpenWork Yes

Saving a Workflow requires a successful tested run, and execution rejects input or output that fails its defined schema. These checks cover saved scripts rather than acceptance criteria on every chat task.

For software work

Hermes Agent and OpenWork compared, with the source and check date of each answer
What you compare Hermes Agent OpenWork
Code review rules Hermes Agent Yes

The pull-request review guide uses reusable review guidelines and remembered team conventions, such as required type annotations and database access rules, to check later diffs.

OpenWork Yes

OpenWork uses OpenCode's configuration, whose custom reviewer can review changes with project instructions included. Review prompts and permissions are configured by you rather than discovered as a separate repository rule system.

Adversarial review Hermes Agent Yes

A bundled pre-commit skill delegates security and logic review to a separate, diff-only reviewer in interactive sessions. Kanban's review skill independently verifies handoffs and keeps the reviewer from editing the implementation.

OpenWork Partly

Its OpenCode engine supports a reviewer subagent with fresh context and its own permissions. The published setup does not establish an independent second review specifically for risky changes.

Risk rating Hermes Agent Not published

Its docs describe no change-level risk rating before merge. Optional residual-risk notes on Kanban handoffs and severity labels for individual review findings do not establish such a rating.

OpenWork Not published

Its public docs do not describe a risk rating for a code change before merge. Model routing for code review and configurable reviewers do not establish such a rating.

Pull requests Hermes Agent Partly

Agents can open pull requests through the GitHub CLI, and the bundled GitHub skill requires reading PR comments. The cited docs do not establish a review-feedback revision cycle.

OpenWork Partly

OpenCode separately documents opening pull requests and responding to line-specific review comments through GitHub Actions. OpenWork's docs do not establish that delivery loop inside its desktop or Cloud chats.

A fact we cannot find or cannot check is left out, and prices can change after the date shown. Task Machine publishes this comparison and is an interested party. Hermes Agent and OpenWork are named only to identify them, and there is no relationship between the companies. Has something changed? Tell us.

Who each one suits

Pick Hermes Agent if you want named specialists, a reviewing orchestrator and editable memory on a profile-based board. Its honest limit for this decision is that workflow-template routing remains unimplemented in the documented kernel.

Pick OpenWork if you want to save a successful script as a versioned Workflow with schema validation. Its honest limit is that those script checks do not establish acceptance checks for every chat task, and its Enterprise permission rollout needs to be checked separately.

Where Task Machine fits

Task Machine addresses this choice through owned Tasks and defined Workflows. A person or agent owns the Task, while a Workflow repeats steps with branches, human questions, approvals and verifier feedback. That fits a report needing both agent work and a recorded human decision.

Task Machine is the wrong pick if your requirement is free agent software and the existing card or script model already fits. Compare Task Machine vs Hermes Agent and Task Machine vs OpenWork against one recurring job, then write down the evidence its final review must require.