Cofounder vs Dock: Company Services or Specialist Handoffs?
Compare Cofounder’s company services with Dock’s specialist handoffs, including scheduled work, approval boundaries and pricing.
Founder, Task Machine
Cofounder and Dock pose a concrete choice: should your existing agent gain company services, or should a Chief of Staff coordinate specialists for you? Cofounder connects an agent you already use to company tools. Dock makes delegation between specialists part of its published operating model.
That distinction matters when a weekly research job needs another owner to finish it. Access to customer records and hosting solves a different problem from briefing a specialist, transferring responsibility and escalating a decision. Start by identifying which handoff you currently perform yourself.
Cofounder gives an existing agent company tools
Cofounder’s introduction describes access through Model Context Protocol (MCP), a command-line interface or an API. Your agent can work with services such as email, customer records, repositories, hosting and databases. This fits a builder who wants to keep an existing agent while giving it company context and operational tools.
Its Tasks guide describes work assigned to a person or Cofounder, with notes, conversations and links. Tasks and larger goal plans are available to selected companies. Check access before making that assignment model part of your team's process.
The company Library gives agents shared files to consult, including customer notes and recorded decisions. Its published model makes those documents editable by users. The guides do not establish dedicated lead roles responsible for coordinating goals, projects and teams.
Dock makes delegation a working structure
Dock’s handoff guide gives the Chief of Staff responsibility for briefing specialists, setting routines and maintaining shared knowledge. Handoffs pass responsibility for finishing work to an agent or person. Agents can escalate decisions through their reporting line.
That structure suits research, sales or operations work that needs several specialists and a clear transfer of responsibility. Shared docs, tables and channels provide places for collaboration. An append-only log records who acted and when.
Keep the release status in the decision: Dock’s terms say it is in beta and does not yet offer a formal uptime service-level agreement. A team that needs a contractual uptime commitment should settle that requirement before moving recurring work.
A reminder and an automated start differ
Cofounder describes scheduled app jobs and scheduled social publishing, but a task follow-up date does not start execution. Its Tasks guide tells you to ask your agent to check due work. The public guides do not describe starting work automatically from outside events or webhooks, even though supported hosts can receive notifications.
Dock’s routine documentation describes hourly, daily, weekly and one-off starts in your timezone. Its webhook guide also explains rules that wake selected agents from incoming events, with optional batching. Those capabilities fit a job that should begin when an event arrives without another instruction from you.
Local availability changes that promise. Dock requires its Mac app to stay open for local routines, and skips an occurrence if the Mac wakes more than 15 minutes after it was due. Cloud routines do not depend on your laptop, so decide where a deadline-sensitive job will run before testing it.
Check what an approval actually protects
Cofounder’s settings documentation includes Ask before going live, requiring administrator approval for staging-to-live changes. Agent guides also require permission before buying, provisioning or moving money, except for free managed setup. Home gathers requests for review and approval.
Dock’s permission controls apply Allow, Ask or Deny to action types for each agent. Ask requires a decision every time. Its Inbox lets you answer work needing a decision without opening the workspace.
For a publishing routine, write down the exact action you want held: producing a draft, publishing it or buying distribution. Cofounder’s published deployment approval and Dock’s per-agent action permissions describe different control boundaries. Neither vendor’s documentation describes evidence-based proposals to revise permissions from your approval history.
Compare the evidence behind a finished result
Cofounder’s verifiers create or improve deliverables and return them after rubric checks succeed. Access remains limited during rollout. Its product implementation checks exclude private workflows and live integrations that the verifier cannot exercise.
Dock describes agents checking their own work and a nightly Chief of Staff review that retains lessons. Its public pages do not describe evaluating delivered results against defined criteria or tests. That is a documentation limit, not evidence that no agent ever tests an output.
Bring an example deliverable to the evaluation, such as a research report with required fields and supporting sources. Ask what check would reject an incomplete report and where you would see that rejection. Cofounder’s rubric-based checks and Dock’s self-review give different published answers to that question.
Cofounder and Dock, point by point
The table presents both products’ answers to the same questions. Each answer includes its source and check date so you can inspect the boundary behind the label.
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 | Cofounder | Dock |
|---|---|---|
| Goals |
Cofounder
Partly
The roadmap tracks milestones and recommends the next one. Larger goal plans and their action tasks are available to selected companies, but the docs do not describe an agent owner tracking a goal's measurable metrics. |
Dock
Partly
Agents accept goals and delegate parts to specialists, and a channel example shows progress toward a lead target. Its docs do not describe tracked goal metrics beyond conversational progress reports. |
| Leads |
Cofounder
Partly
Agents read shared company context, recent events and the recommended roadmap milestone. The docs do not describe lead or owner roles coordinating work across goals, projects or teams. |
Dock
Yes
A Chief of Staff runs work involving several agents, briefs each specialist, sets routines and keeps shared knowledge current. Agents escalate decisions up their reporting line. |
| Tasks |
Cofounder
Yes
Selected companies can assign tasks to Cofounder or a person and keep notes, conversations and links. |
Dock
Yes
Handoffs transfer responsibility for finishing a task to an agent or a person. An append-only log records who did what and when. |
| Recurring work |
Cofounder
Partly
Cofounder describes scheduled jobs for your app and scheduled social posts. A task follow-up date does not schedule work automatically. Its docs do not describe starting work from outside events or webhooks. |
Dock
Yes
Agents run hourly, daily, weekly or one-off routines in your timezone. Webhook rules wake specific agents from outside events, with optional batching between check-ins. |
| Workflows |
Cofounder
Partly
Cofounder lists scheduled jobs and workflows for your app. Its public guides do not describe defining repeatable multi-step processes with steps, branches or conditions. |
Dock
Partly
You can split recurring work among specialists and specify handoffs and human decisions. Dock describes agents adapting their steps, rather than a defined multi-step process that runs the same way each time. |
| Autonomy |
Cofounder
Yes
Ask before going live requires an administrator to approve changes moving from staging to live, and requested reviews and approvals are collected on Home. Agent guides also call for approval before buying, provisioning or moving money, with an exception for free managed setup. |
Dock
Yes
Each agent has Allow, Ask or Deny permissions by action type, with Ask requiring approval every time. The Inbox collects work needing a decision and lets you reply without opening the workspace. |
| Learned autonomy |
Cofounder
Not published
Agents follow guides about approvals and report progress. The docs describe no evidence-based proposals to change an agent's autonomy from approval history. |
Dock
Not published
Its docs describe manually moving permissions from Ask to Allow and learning your working preferences. They do not describe proposals to change an agent's authority based on your approval history. |
| Memory |
Cofounder
Yes
The company Library holds shared files such as customer notes and decisions, which agents read for later work and users can read and edit. The docs do not describe automatic lessons from corrections or restorable notes across agent and project scopes. |
Dock
Yes
Agent memory updates during work and carries standing knowledge into later conversations. You can read, edit or delete entries in the Memory tab. |
| Planning |
Cofounder
Yes
An agent suggests tasks from the roadmap and waits for you to pick. Selected companies can keep larger goal plans in the Library and create tasks for concrete actions. The docs do not describe reviewing acceptance criteria before every run. |
Dock
Partly
The Chief of Staff plans and coordinates work in conversation, and shared docs can hold plans. Its docs do not establish a conversation becoming a plan artifact reviewed before execution. |
| Intake |
Cofounder
Partly
Supported hosts can receive webhook notifications for incoming CRM emails, texts, forms and social interactions. The docs do not describe routing those incoming requests to an owner or agent. |
Dock
Yes
Each connected tool gets a webhook URL, and plain-language rules route events to agents. Channel events can also tag the right person and let an agent pick up work in the thread. |
| Connectors |
Cofounder
Yes
Your agent uses Cofounder through MCP, a command-line interface or an API, and supported hosts can subscribe to event notifications. The docs do not describe installing arbitrary services as custom connectors that start workflows. |
Dock
Yes
The official CLI documents MCP endpoints, API keys and webhook configuration. Incoming webhook rules wake agents from connected services. |
| Models |
Cofounder
Partly
You connect your own agent, including Claude, ChatGPT, Codex or Cursor. Cofounder does not document automatic model selection for each task or step. |
Dock
Partly
You choose a model and reasoning effort for each agent and can switch models in the composer. Its docs do not describe automatic model selection for each task or step. |
| Cloud or your own computer |
Cofounder
Yes
Cofounder sandboxes let your agent edit files and run code on web and mobile. On desktop, it can use a sandbox or the terminal in your desktop app. |
Dock
Yes
Choose cloud agents on Dock's machines or local agents through the Mac app. Cloud agents keep working with your laptop shut, while local agents can access your Mac's files and tools. |
| Your own subscription |
Cofounder
Yes
Cofounder connects to your own Claude Code or Codex agent through MCP. Your agent provider's subscription or model usage charges are separate from Cofounder's service credits. |
Dock
Partly
Connected Claude subscriptions and supported ChatGPT models can fund local agents instead of Dock usage. Cloud agents always use Dock usage, and not every OpenAI model supports subscription funding. |
| Pricing |
Cofounder
Yes
Plans start at $40 a month, all as usage credit, and unused credit rolls over. Google and Meta ads cost ad spend plus 10%. |
Dock
Yes
Starter is $40, Standard $100 and Scale $200 per person per month, including 5, 10 or 20 cloud agents and 10,000, 20,000 or 50,000 business credits. Additional usage or credits can be purchased and roll over. |
| Correctness |
Cofounder
Yes
Limited-rollout verifiers create or improve deliverables and return them after rubric checks pass. Product implementation checks exclude private workflows and live integrations the verifier cannot exercise. |
Dock
Partly
Dock says agents check their own results, and the Chief of Staff runs a nightly self-review to retain lessons. Its docs do not describe evaluating delivered work against defined criteria or tests. |
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. Cofounder and Dock are named only to identify them, and there is no relationship between the companies. Has something changed? Tell us.
Who each one suits
Pick Cofounder if you want your existing agent connected to company tools and managed services. Its pricing documentation lists plans from $40 a month, all as usage credit, with unused credit rolling over. Google and Meta ads cost ad spend plus 10%. Tasks and verifiers have rollout restrictions, so confirm access to the features your process needs.
Pick Dock if a Chief of Staff, specialist handoffs and event-driven routines match your team’s work. Its pricing page lists Starter at $40 per person per month, including 5 cloud agents and 10,000 business credits. Additional usage or credits can be purchased and roll over. Dock remains in beta, and local routines depend on the Mac app being available.
Where Task Machine fits
Task Machine addresses a different part of the handoff: a person or agent owns a Task containing its plan, decisions and run history. Its Workflow documentation describes defined steps, branches, human approvals and verifier checks that return failed work with feedback. That fits recurring work whose route and acceptance checks need to stay explicit.
Task Machine is the wrong pick if you need managed company hosting and databases, since its docs do not describe a managed product stack that provisions your business hosting and database together. See Switch from Cofounder to Task Machine. Read Task Machine vs Cofounder and Task Machine vs Dock with one weekly job in hand, then mark where that job needs an owner transfer, an approval and a result check.