Codex vs OpenWork: Repository Work or Shared Agent Tools?

6 min read Comparisons

Compare Codex and OpenWork on issue intake, reusable processes, model choice and the controls available to a small team.

Codex and OpenWork raise a work-intake question: should an agent take an assigned repository issue, or answer requests through an individual's assistant session? Codex's Linear integration accepts issue assignments and posts progress back. OpenWork's Slack bot documentation describes requests reaching each invoking member's assistant, with separate sessions even inside the same thread.

That distinction matters when a small team shares tools but still needs to know who owns a request. OpenWork's bot is private alpha for approved organizations. Codex's published issue flow gives you an issue and a continuing cloud chat, which is a different coordination model from member-specific service access.

Codex anchors the agent to code and issues

Codex plans, edits and reviews code through its app, terminal, editor and cloud workspaces. Its Goal mode guide lets you keep a persistent target with measurable completion criteria. For a repair, that target can be the test or measurement you will use to accept the result.

Linear assignments, mentions and triage rules can start cloud work. Codex selects a repository environment and posts progress to the issue, and follow-up comments continue that cloud chat. This fits a team whose work already arrives as repository changes with an issue to track the request.

The limit is the surrounding process definition. Codex's skills and plugins documentation packages reusable instructions, tools and runtime hooks. It does not establish a fixed execution graph with enforced branches and steps, so assess what your team still needs to enforce around the session.

OpenWork shares the assistant's working tools

OpenWork runs the OpenCode engine through desktop and browser apps. Its desktop page describes local work with your provider key or local model. Team plans can distribute provider keys through the organization, giving members access to a shared model setup.

OpenWork also lets you turn a completed chat into a reusable skill, with editable instructions for later work. Its memory documentation describes searching saved sessions for recall. Searches depend on the current app and workspace context, and transcript reads return recent messages, so older sections of a long conversation may not appear.

Work stays in chats that members can name, group, archive and restore. Its public pages do not describe work records assignable to both people and agents. Sharing tools and recalling a conversation can help a team, but neither establishes who must finish an unresolved deliverable.

Instructions, saved scripts and result checks

Codex skills give an agent process guidance. Goal mode can then test progress against measurements and completion criteria, while cloud work exposes changed files and check results for inspection. That makes a concrete coding outcome the center of the review.

OpenWork's Workflow API describes execution of an exact saved version with supplied inputs and live tools. A successful tested Code Mode script can be saved as a Workflow, and execution validates input and output schemas while recording a receipt.

For a repeated script with a defined data shape, versioned execution and schema checks are useful evidence. They cover saved scripts, however, and do not establish acceptance criteria for every OpenWork chat task. Compare a schema failure with a semantically wrong result that still has the right shape.

A fair trial should include both cases. Ask where each failure is recorded and what the reviewer needs to decide. A successful script run and a fulfilled request are separate claims, even when the same execution contributes evidence for both.

Permission scope and model selection

Codex's approval guide covers sandbox settings, command rules and configured approval policies. OpenWork's Agent permissions guide describes Enterprise controls to allow, ask first or block commands, file edits and websites. Those OpenWork controls are rolling out and require the OpenCode v2 engine.

A team evaluating central rules should confirm availability before assuming the documented settings are active in its installation. OpenWork says desktop apps keep their latest cached rules when Den is unreachable. That matters if an administrator expects a new restriction to take effect immediately.

Model selection differs too. Codex publishes manual model and reasoning-effort choices. OpenWork's Smart Routers choose a model and thinking level for a prompt from administrator-selected models, then keep them for following tool steps until the task ends.

Read the add-ons before choosing an environment

Codex's pricing documentation lists USD prices of $20 a month for Plus and Pro from $100 a month. Local messages and cloud chats share the plan allowance. API-key usage has separate API billing.

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. On Team and Enterprise, Cloud computers add $50 per member monthly and models $10 per user monthly.

OpenWork's browser computer is a separate paid add-on, not an execution environment implied by every free desktop installation. Its Cloud Automations can continue when your computer is off, while desktop Automations need the app open and signed in. Decide which computer must execute the job before comparing the account plans.

Codex and OpenWork, point by point

The table compares the same questions with a source and check date for each answer. The descriptions identify private-alpha intake and rolling permissions separately from generally described capabilities.

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.

Codex and OpenWork compared, with the source and check date of each answer
What you compare Codex OpenWork
Goals Codex Yes

Goal mode gives Codex a persistent target to track while a task runs. You can define tests or measurements as completion criteria and steer the same session as it works toward them.

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 Codex Partly

A main agent delegates to specialized subagents, routes follow-up instructions and collects their results. Its docs describe coordination within a chat rather than assigned lead roles over projects, goals or teams.

learn.chatgpt.com/…/subagents.md
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 Codex Yes

Through its Linear integration, you can assign issues to Codex just as you assign them to teammates. Codex posts progress and results on the issue, and follow-up comments continue the same cloud chat.

learn.chatgpt.com/…/linear.md
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 Codex Yes

Codex in the desktop app can run scheduled tasks in local projects or isolated worktrees. Separately, Linear triage rules can automatically delegate new issues to Codex cloud chats. Local schedules need the computer and app running.

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 Codex Partly

Skills package reusable process instructions and resources, and plugins add tools and runtime hooks. Its docs describe agent-followed instructions rather than a fixed execution graph with enforced steps and branches.

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 Codex Yes

Sandbox settings, command rules and approval policies determine when Codex asks before acting. Requests route to you by default, with desktop permission notifications and an Activity view for chats waiting for a response when available.

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 Codex Not published

Its docs do not describe Codex proposing permission changes from your approval history. Auto-review evaluates individual action requests under a configured policy, while local memories carry context into later work.

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 Codex Yes

Opt-in local memories turn eligible earlier chats into files that inform later sessions. You can inspect the files, control each chat's use and contribution, or delete the store. The docs discourage hand editing as the primary control.

learn.chatgpt.com/…/memories.md
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 Codex Yes

Plan mode lets you ask Codex for an execution plan before implementation starts. You can refine an unclear outcome through conversation into a goal with measurable success criteria before starting Goal mode.

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 Codex Yes

Linear assignments, mentions and triage rules route issue work to Codex. It selects a repository environment, starts a cloud chat and posts progress back to the issue.

learn.chatgpt.com/…/linear.md
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 Codex Yes

Cloud environments connect Codex to APIs and private services. Local clients can connect Linear through Model Context Protocol, and Linear triage rules can automatically start Codex cloud work.

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 Codex Partly

You choose the model and reasoning effort through app controls, the CLI or the editor. Its docs describe manual choices and defaults rather than automatic model selection for each task or step.

learn.chatgpt.com/docs/models.md
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 Codex Yes

The CLI works against repositories and tools on your own machine. Codex Cloud provides a separate workspace for each task and keeps working while your computer sleeps.

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 Codex Yes

ChatGPT sign-in covers subscription access, and an OpenAI API key provides usage-based access for local app, CLI and editor work. API-key access does not include Codex cloud, which requires ChatGPT sign-in.

learn.chatgpt.com/docs/auth.md
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 Codex Yes

USD pricing lists Plus at $20 a month and Pro from $100 a month, with included usage and extra credits. API-key usage is billed separately at API rates. Local messages and cloud chats share the plan allowance.

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 Codex Yes

Goal mode can verify progress against tests, measurements and completion criteria. Codex also offers dedicated code review and cloud tasks whose changed files and check results you inspect before accepting the work.

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

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

Codex reviews GitHub pull requests against applicable code review rules in AGENTS.md files. Root guidance and more specific nested guidance apply to the changed files they cover.

learn.chatgpt.com/…/github.md
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 Codex Yes

Security Review adds a separate security-focused pass over pull requests alongside Code Review, using threat-model context when configured. It is a research preview for Pro, Business, Enterprise and Edu, not Plus.

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 Codex Partly

Auto-review assesses the risk and authorization of individual proposed actions. Its docs describe action-level risk categories rather than an overall risk rating for a code change before merge.

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 Codex Yes

Codex cloud tasks can open pull requests, and authors can ask Codex to address specific review feedback. Its review interface also exposes checks and lets you investigate failing checks before deciding to merge.

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. Codex 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 Codex if issue assignment, measurable coding goals and reviewable repository changes define your team's work. Reusable skills are agent-followed instructions, so a defined branching execution process needs separate consideration.

Pick OpenWork if members need shared tools, model setup and reusable scripts alongside their chats. Confirm private-alpha bot access and rolling Enterprise permissions before relying on either, and keep responsibility for unresolved work explicit.

Where Task Machine fits

Task Machine focuses on the ownership question behind this comparison. Its Tasks can belong to people or agents and retain plans, decisions, discussion and run history. Task Machine also brings consequential decisions to the Inbox and checks delivered work against acceptance requirements, making a handoff more specific than reopening a chat.

Task Machine is the wrong pick if you primarily want OpenWork's desktop access and shared OpenCode tools or Codex's repository-adjacent coding session. Use Task Machine vs Codex and Task Machine vs OpenWork to map one unresolved request to its owner, acceptance requirements and next reviewer.