---
title: "Dock + ClickUp: agent-drafted project briefs with attributed stakeholder review"
excerpt: "Dock turns a ClickUp task hierarchy into an agent-drafted project brief, with every scope decision attributed to the agent that wrote it and the stakeholder who signed off."
author: mei
category: Use Cases
date: "2026-05-30"
---

A ClickUp space holds the task tree. Dock holds the project brief the agent wrote from it. The agent reads the parent task, subtasks, dependencies, and custom fields, then drafts a brief: scope, deliverables, owners, risks. Stakeholders approve in Dock, and the agent reconciles changes back to ClickUp. ClickUp stays the plan of record. Dock holds the interpretation.

ClickUp and Notion stay the system of record for the raw data. Dock is the system of record for what the AGENT INTERPRETS. Each Dock row carries a pointer back to the platform record, agent identity, decision, reviewer, and timestamp. The agent re-fetches platform data via fresh API reads when it needs current state.

## The surface: a Project Brief table

| brief_id | clickup_task_url | scope_statement | deliverables | risk_flag | drafted_by | reviewer | signed_off_at |
|---|---|---|---|---|---|---|---|
| PB-118 | clickup.com/t/86c1a2 | Migrate billing module to event-driven queue by Q3 | Queue infra, dual-write shim, cutover runbook | Dependency on platform team Q2 ship | agent:mei-pm | govind@dock | 2026-05-29 14:02 |
| PB-119 | clickup.com/t/86c1b7 | Ship self-serve seat management to mid-market tier | Admin UI, billing webhook, in-app tour | Pricing model not finalized | agent:mei-pm | sarah@dock | pending |
| PB-120 | clickup.com/t/86c1d4 | Consolidate three onboarding flows into one | Flow audit, merged spec, removal plan for legacy | Two flows owned by different squads | agent:mei-pm | not assigned | pending |

Every row points back to a ClickUp task. The `drafted_by` column names the agent. The `reviewer` and `signed_off_at` columns make the human sign-off auditable. The agent does not edit the ClickUp task tree until a brief is signed off.

## One worked workflow

A product lead opens a ClickUp parent task with twelve subtasks. The agent reads the tree, pulls custom fields for effort, and notices three subtasks share a dependency on another squad. It drafts PB-119, flags the dependency in `risk_flag`, and tags the squad lead in a Dock comment. The squad lead replies that the dependency ships two weeks later than the brief assumes. The agent updates deliverables and marks the brief ready. Sarah reviews, approves with one wording change, and signs. The agent then edits the affected ClickUp subtasks and links each edit back to PB-119.

## Why this matters

Briefs that live only in a doc tool drift from the task tree within a sprint. Briefs that live only in the task tool lose the reasoning behind scope choices. Dock holds the reasoning and attributes it. See [Dock for project management](/blog/dock-for-project-management) and the [agent collaboration primer](/blog/agent-collaboration-primer) for how multiple agents share one brief table. The pattern follows the broader [Cloud 2.0 for product](/blog/cloud-2-0-for-product) thesis.

ClickUp documents that subtasks and dependencies "break big work into trackable, manageable pieces"<sup>[1](https://clickup.com/features/tasks)</sup>. ClickUp Docs are "beautiful, collaborative docs right where you work"<sup>[2](https://clickup.com/features/docs)</sup>, but they do not separate agent-drafted from human-edited content and do not carry agent identity. Dock adds that layer.

Set up a Project Brief surface and connect your ClickUp space to try it.

## FAQ

**Why not write the brief directly in ClickUp Docs?**
ClickUp Docs do not distinguish agent-drafted text from human-edited text, and they do not attach an agent identity to a draft. Dock rows carry [agent identity](/blog/agent-identity) on every write, so a stakeholder always knows which agent wrote a scope statement and which human signed off on it.

**Does the agent overwrite my ClickUp task tree?**
No. The agent only edits ClickUp tasks after a brief is signed off in Dock, and every edit is logged against the signed brief. See [agent identity lifecycle](/blog/agent-identity-lifecycle) for how the agent's write scopes are bounded.

**What happens when ClickUp data changes after sign-off?**
The agent re-fetches the ClickUp tree on a schedule. If a source task moves or a dependency changes, the agent posts a drift notice on the brief row and waits for a human to decide whether to re-open sign-off.

**Can two agents draft briefs against the same ClickUp space?**
Yes. Each draft is attributed to the agent that wrote it, and the brief table enforces one active draft per parent task. The [Cloud 2.0 for product](/blog/cloud-2-0-for-product) post covers the broader pattern.