Devin vs Herdr: Delegated Sessions or Your Own Terminals?
Compare Devin engineering sessions with Herdr terminal supervision, including machine control, approvals, recurring work and review evidence.
Founder, Task Machine
Devin and Herdr put the responsibility for execution in different places: Devin supplies repository-connected sessions and delegated coordination, while Herdr gives you a desktop interface for coding-agent terminals on machines you operate. The decision is whether you need a delivery process to delegate into or direct control of the agent programs already on your computers.
Both can arrange helpers and run tests. That overlap does not make their control boundaries equivalent. Before choosing, decide who supplies the machines, who enforces permissions and where you will inspect the evidence that a change is ready.
Devin connects a request to engineering delivery
Devin's advanced-capabilities documentation describes a coordinator that scopes a large job, delegates to managed sessions and resolves conflicts. That supports work spread across several implementation sessions. The coordinator is responsible for the delegated project, not a permanent team lead role.
A request can begin in Linear, where assigning a ticket to Devin starts work with an activity feed, synchronized plan and follow-up messages. Ask Devin supports a codebase discussion before implementation and turns the conversation into a context-rich prompt. These paths suit a team that wants scoping and progress attached to an engineering request.
Its review and browser-testing capabilities also belong in the evaluation. Devin can build a focused test plan, execute it and return a recording. You can judge a change using both the code and an observable test artifact.
Herdr exposes the terminals you already run
Herdr's core documentation describes a runtime for starting agent terminals, sending work and receiving reports. A parent agent can start helpers, wait for them and read their results. Its interface keeps those programs accessible as terminals you can inspect and answer.
The Herdr guide suggests a space per repository and a tab per task, with worktrees for issue work. It also describes editable layouts and mods. That suits a developer whose main coordination problem is seeing several coding agents and reaching whichever terminal needs attention.
Its official mobile app connects to herdr-core over SSH and lets you answer prompts in the real terminal. The current desktop app and core are the relevant runtime here. The older Herdr TUI has a separate server and does not share agents with the desktop app or official mobile clients.
Machine access is a responsibility boundary
Devin offers cloud sessions and Outposts for command execution, file edits and repository access on your machines. Outposts is available on Pro, Max and Teams. Inference and planning remain in Devin's cloud, so customer-run commands do not mean the entire process runs locally.
Herdr runs agents on your computer or on machines connected through SSH, including servers, virtual machines and cloud sandboxes. Its current public app and runtime docs do not describe Herdr-hosted execution. You supply the terminal agents and use their existing model access and accounts.
Closing the Herdr window does not stop those agents, but rebooting the machine stops the programs. Panes return as shells, and Herdr does not restart agents or resume their work. For an overnight job, include machine restarts in your operating plan before treating persistent terminals as resumable work.
A prompt in an Inbox is not a permission policy
Devin CLI's permissions reference lets you allow, deny or require approval for commands, file access and Model Context Protocol tools. Approval prompts can grant access for one action, a session, a project or globally. That gives you documented scopes for a permission decision.
Herdr's mobile Inbox lists agents needing attention and opens their terminal prompts. Its security documentation explicitly says core is not a permission broker or sandbox. A client connected to its API can run commands and change files with your account's access.
The mobile attention surface helps you reach an agent, but permission enforcement remains a separate concern. Herdr also treats mods as trusted code, so reviewing a customization matters before installing it. Neither product's public docs describe proposing permission changes from approval history.
Repetition and review need their own mechanisms
Devin's Automations guide describes schedules, service events and authenticated webhooks, with conditions and optional preflight checks. Dynamic Workflows use deterministic Python scripts with recorded calls and replay after interruption. For recurring engineering work, those are explicit orchestration mechanisms.
Herdr templates save an agent command and role, and helpers can exchange work and reports. Custom mods can repeat code on an interval, but its public docs do not describe an outside-event or webhook setup for recurring work. They also do not describe a defined multi-step process that executes the same way each time.
Both can run tests, but the result-checking process differs. Herdr can delegate a test command and read the result, with checks arranged through terminal programs. Devin's review guide describes repository instructions and beta custom reviewers, while its beta Autoapprover returns an approval decision with reasoning. Herdr's reviewer templates and separate-reviewer examples should not be described as an automatic gate for risky changes.
Devin and Herdr, point by point
This table asks each product the same questions and includes the source and check date for each answer. The explanations show where a terminal capability is narrower than a delivery or approval process.
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 | Devin | Herdr |
|---|---|---|
| Goals |
Devin
Partly
Devin can coordinate work toward a requested outcome, such as improving modules below a test-coverage threshold. Its docs do not describe an owned goal with a tracked metric and ongoing reviews against its target. |
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 |
Devin
Yes
A coordinator session scopes a large job, delegates to managed Devin sessions, monitors progress, resolves conflicts and compiles results. This coordinates the delegated project rather than establishing a permanent team lead. |
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 |
Devin
Yes
Through Linear, tickets can be assigned to Devin and started with a default skill. Devin posts an activity feed, syncs its plan and accepts follow-up messages on the ticket's agent session. The work record lives in Linear. |
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 |
Devin
Yes
Automations start Devin sessions on hourly, daily, weekly or custom schedules and from outside events, including GitHub checks, Linear updates and authenticated webhooks. |
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 |
Devin
Yes
Dynamic Workflows use deterministic Python scripts to order agents and pass structured results between stages. Calls are recorded, and interrupted runs replay completed results while unfinished work runs again. |
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 |
Devin
Yes
Devin CLI lets you allow, deny or require approval for commands, file access and Model Context Protocol tools. Approval prompts let you grant access once, for a session, for a project or globally. |
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 |
Devin
Not published
Its docs describe memory from corrections and configurable approval rules, but do not describe Devin proposing permission changes from your approval history. |
Herdr
Not published
Its public app and runtime docs do not describe proposing changes to an agent's permissions from your approval history. |
| Memory |
Devin
Yes
Personal memory carries preferences and corrections into later sessions. You can browse notes and ask Devin to change or remove them. Memory is personal within an organization, and automation sessions do not read or write it. |
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 |
Devin
Yes
Ask Devin lets you discuss your codebase and scope implementation before coding. It turns that conversation into a context-rich prompt for an Agent session, whose progress remains visible in the conversation. |
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 |
Devin
Yes
Automations route Slack messages, tickets and webhook events into new Devin sessions, existing sessions or a triage monitor. Conditions and optional preflight checks decide which events start work. |
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 |
Devin
Yes
Devin CLI connects custom Model Context Protocol servers to external tools. Automations offer native service triggers and incoming HTTP webhooks with configurable conditions. |
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 |
Devin
Yes
Adaptive in Devin CLI evaluates each request and selects an underlying model, routing simpler tasks to faster models and complex tasks to more capable ones. You can select a specific model instead. |
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 |
Devin
Yes
Devin offers cloud sessions and Outposts for command execution, file edits and repository access on your own machines. Outposts keeps inference and planning in Devin's cloud and is available on Pro, Max and Teams. |
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 |
Devin
Partly
Pro, Max and Teams can connect personal ChatGPT Go, Plus or Pro subscriptions for eligible GPT requests. Other models and cloud machine time still use Devin funding, and Free and Enterprise accounts are excluded. |
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 |
Devin
Yes
Free is $0. Pro is $20/month and Max $200/month, with usage allowances and purchasable on-demand credits. The pricing page and self-serve guide disagree on whether Teams' $80/month is a base charge or a minimum. |
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 |
Devin
Yes
Devin builds a focused test plan from a code change, executes it through the browser and sends a recording. Beta custom reviewers also check pull requests against written instructions, with configurable passing conditions. |
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 | Devin | Herdr |
|---|---|---|
| Code review rules |
Devin
Yes
Devin Review reads repository instructions such as REVIEW.md and AGENTS.md. Beta custom reviewers check pull requests against your instructions or skills and produce attributed findings. |
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 |
Devin
Partly
Beta custom reviewers can add focused checks alongside Devin Review, for example API compatibility. Its docs do not establish that these are independent second reviews specifically triggered by risky changes. |
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 |
Devin
Partly
Beta Autoapprover evaluates intent, code quality and incident risk before approving a commit or leaving it for human review. Its published result is an approval decision with reasoning, not an overall merge-risk rating. |
Herdr
Not published
Its public app and runtime docs do not describe rating the risk of a code change before merge. |
| Pull requests |
Devin
Yes
Devin creates GitHub pull requests, responds to review comments and pushes branches. Repository permissions control which operations it can perform. |
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. Devin 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 Devin if you want delegated engineering sessions, event-triggered work and review artifacts around repository delivery. Its pricing page lists Free at $0, Pro at $20/month and Max at $200/month with usage allowances and on-demand credits. Confirm Teams terms because the pricing page and self-serve guide disagree about whether $80/month is a base charge or a minimum, and remember that automation sessions do not read or write personal memory.
Pick Herdr if you want direct supervision of existing terminal agents on local and SSH-connected machines. Its homepage describes the free, open-source app and runtime, and the official mobile app is free too. Its public pages do not describe an included model allowance or Herdr-hosted execution, and you remain responsible for machine availability and access controls.
Where Task Machine fits
Task Machine fits a different responsibility boundary: people and agents own the same kind of Task, and consequential decisions arrive with context in the Inbox. Its Workflows define repeated steps, approvals and verifier checks. It is the wrong pick when you mainly want an editable terminal desktop or when Devin's repository delivery already supplies the process you need.
Use Task Machine vs Devin and Task Machine vs Herdr to inspect the work-record and review differences. Choose one overnight job and record its machine owner, permission policy and required completion evidence before comparing those rows.