Codex vs Herdr: Coding Sessions or Terminal Supervision?
Compare Codex and Herdr on execution, approvals and recurring checks to decide who should operate your coding environment.
Founder, Task Machine
Choosing between Codex and Herdr starts with a practical question: do you want a coding agent with hosted execution, or a desktop interface for supervising terminal agents you already run? Codex's execution documentation describes a separate cloud workspace for each task. Herdr's runtime overview centers on agent terminals on machines you control.
That distinction decides who operates the environment for a long-running repository change. Codex can continue a cloud task while your computer sleeps. Herdr connects your computers and remote machines, which suits developers who want direct access to the programs and terminals doing the work.
Codex keeps implementation and review together
Codex supports planning, editing and reviewing code across its app, terminal, editor and cloud workspaces. Its Goal mode guide describes a persistent target with tests or measurements as completion criteria. You can refine the target and steer the running session as it works toward that outcome.
For an issue-driven team, its Linear integration gives the session an existing work record. Assign an issue to Codex and it posts progress and results there. Follow-up comments continue the same cloud chat, so a request to revise the implementation can stay attached to the issue that started it.
This fits a developer who wants to move from a proposed fix to a reviewable result without operating a separate terminal-management interface. It still leaves you responsible for deciding whether the changes meet the intended outcome. A completed session and an acceptable patch are different decisions.
Herdr makes your terminals visible
Herdr's desktop overview describes spaces, tabs and real coding-agent terminals. Its suggested arrangement is a space for a repository and a tab for a task. An agent can work in an issue's worktree, launch a helper and receive that helper's report.
The benefit is direct supervision of programs you already use. Herdr runs existing terminal agents unchanged, using those programs' accounts and model access. Remote access uses SSH, so the same supervision model extends to a server or virtual machine you connect.
Its mobile app groups agents by whether they need attention, are done, working or idle. Opening an agent lets you answer its real terminal prompt. Its public pages do not describe shared work records that both people and agents can own with their own discussion or history.
A permission request is different from a terminal prompt
Codex's approval documentation describes sandbox settings, command rules and approval policies. These controls determine when the agent needs your permission before acting.
Herdr's mobile Inbox helps you find and answer agents that need a response. Its security documentation explicitly says core is not a sandbox or permission broker. Commands run with your account's access, and a client connected to the API can execute commands and change files with that access.
A terminal waiting for your input therefore does not establish an action-approval policy in Herdr itself. If you choose Herdr, inspect the controls of each installed agent and the access you give connected clients. If you choose Codex, review its configured policy before treating unattended execution as permission to perform every proposed action.
Repeating a check and judging its result
Codex's automation guide covers scheduled work in local projects or isolated worktrees. Local schedules require both the computer and app to remain running. Linear triage rules offer a separate route for new issues to start cloud work.
Herdr publishes an interval timer for custom mods in its mod instructions. Its public pages do not describe an outside-event or webhook setup that starts recurring work. An interval-based custom check and issue-triggered cloud delegation solve different intake problems.
For result checks, Codex can use Goal mode's completion criteria and exposes cloud changes and check results for inspection. Herdr can run test commands and delegate a test run to a helper, then return the report. Those Herdr checks are arrangements you build with terminal programs, not a published Herdr acceptance-review gate.
Use the same repository check when evaluating both. Specify the expected evidence, where you will find a failure and who must respond. Merely seeing an agent stop working should not be your success criterion.
Codex and Herdr, point by point
The table presents both products' answers to the same questions. Each answer carries its source and check date, including the limits of what the public documentation establishes.
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.
| What you compare | Codex | Herdr |
|---|---|---|
| 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. |
Herdr
Partly
A community Claude goal mod writes a /goal for your current Claude Code session. Its listing does not describe tracking a measurable target and reviewing progress against it, and Herdr does not review gallery mods. |
| 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. |
Herdr
Partly
A parent agent starts helpers, gives them work and receives their reports. Its docs do not describe a project, goal or team lead role that reviews and coordinates the 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. |
Herdr
Partly
Herdr suggests a tab per task and can start an agent in a worktree for an issue. Its docs do not describe shared work records that both people and agents can own with their own discussion or history. |
| 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. |
Herdr
Partly
Custom mods can repeat code on an interval with cx.timer.every. Its docs do not describe an outside-event or webhook setup that starts recurring work. |
| 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. |
Herdr
Partly
Templates save an agent command and role, and agents can start helpers, wait and read their results. Its docs do not describe a defined multi-step process that runs the same way each time. |
| 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. |
Herdr
Partly
The mobile Inbox lists agents needing you and lets you answer their terminal prompts. Core explicitly is not a permission broker for agents, and its docs do not describe configurable action approvals in Herdr itself. |
| 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. |
Herdr
Not published
Its public app and runtime docs do not describe proposing changes to an agent's permissions from your approval history. |
| 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. |
Herdr
Not published
Core retains terminal history and opaque API documents. Its docs do not describe agent memories that improve later work and can be inspected or corrected. |
| 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. |
Herdr
Partly
The community Plans layout displays repository plan files, tickets and the agents working on them. Its listing does not establish planning in conversation before execution, and Herdr does not review gallery layouts. |
| 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. |
Herdr
Yes
Custom clients can use core's API to start or message agent terminals. The caller selects the terminal or agent template that receives the work, rather than sending it to a shared business-request queue. |
| 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. |
Herdr
Partly
Mods can fetch URLs, and the guide shows a custom ntfy notification. Its docs do not describe MCP connectors or outside-event triggers alongside custom HTTP access. |
| 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. |
Herdr
Partly
Agent templates put model choices in the command that launches your installed agent. Its docs do not describe Herdr selecting a model automatically per task or step. |
| 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. |
Herdr
Partly
Agents run on your computer or machines you connect over SSH, including servers, virtual machines and cloud sandboxes. Its current app and runtime docs do not describe a Herdr-hosted execution option. |
| 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. |
Herdr
Yes
Herdr runs your existing terminal agents unchanged, including Claude Code. You use those programs' own model access and accounts rather than a Herdr model subscription. |
| 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. |
Herdr
Partly
The app and herdr-core are free and open source, and the official mobile app is free too. Its pages do not publish usage-credit pricing or an included model allowance. |
| 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. |
Herdr
Yes
Core runs test commands such as cargo test, and an agent can delegate a test run and read its result. These are checks you arrange through terminal programs rather than a Herdr acceptance-review gate. |
For software work
| What you compare | Codex | Herdr |
|---|---|---|
| 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. |
Herdr
Partly
A saved reviewer template instructs an agent to read code and report bugs with file paths, lines and reasons. Its docs do not describe reviewing code against configured repository rules. |
| 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. |
Herdr
Partly
An agent can hand a diff to a separate reviewer, and the guide shows a reviewer reporting back. Its docs do not establish a distinct independent review 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. |
Herdr
Not published
Its public app and runtime docs do not describe rating the risk of a code change before merge. |
| 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. |
Herdr
Partly
The community PR Reviews mod checks out a selected pull request in a worktree and starts your agent with a configured prompt. Its listing does not establish agents addressing review feedback, and Herdr does not review gallery mods. |
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 Herdr 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 you want a coding session with measurable completion criteria, issue delegation and a hosted workspace. Its pricing documentation lists Plus at $20 a month and Pro from $100 a month in USD. Local messages and cloud chats share the plan allowance, and API-key usage has separate API billing. Local scheduled jobs still depend on your computer and app staying available.
Pick Herdr if supervising existing agents across local and remote terminals is your main job. Its app and herdr-core are free and open source, and the official mobile app is free. That software does not establish an included model or hosting allowance, and core leaves permission enforcement to other controls.
Where Task Machine fits
Task Machine addresses the shared-work boundary in this decision. Its Task documentation describes records owned by a person or agent, keeping the plan, discussion, decisions and runs together. Task Machine also provides Workflows with verifier feedback, so a repeated check can send a failed result back for revision before the work moves on.
Task Machine is the wrong pick if you primarily want an editable desktop interface for existing terminal agents, or a coding agent beside your source. Compare Task Machine vs Codex and Task Machine vs Herdr against one recurring repository check, and identify who should own the failed result before choosing.