Getting the Most Out of Claude Cowork: 7 Tips for PMs | PM Toolkit

I have used Claude Cowork since its January 2026 launch for morning briefs, research across Slack and email, and weekly rollups. It took a few weeks to make those outputs useful. I spent about 30 minutes configuring the environment, then adjusted it over time. The steps below cover the context files and instructions I use.

1. Put a _Context.md in every working folder

The underscore prefix places the file near the top of an alphabetical file list. Inside, I identify which files are sources of truth and which to skip. That distinction matters when a folder has 40 files but only 5 are relevant: old brainstorm notes should not be treated as current strategy. I use Cowork to help keep these instructions updated.

My _Context.md for a typical project folder looks like this:

# Context
 
## Read these
- product-roadmap-h1.md
- project-brief.md
 
## Skip these
- old-drafts/
- archive/
- meeting-notes-jan.md (outdated)

The filename is my convention, not a special file Cowork automatically reads. I tell it to check the file in my instructions.

2. Add a personal context file

Separate from the project context, I have a file that tells Cowork who I am. It covers my name, role, team, responsibilities, and preferred output structure.

My file includes:

My role and what I own
I prefer tables over paragraphs when comparing options
Keep language simple, no filler phrases
Always cite source filenames when synthesizing research so I can trace where a claim came from

Keep this file somewhere Cowork can access, and tell it to read the file in your global instructions. Update it when your role or preferences change.

3. Require a plan before execution

This is a simple rule I added to my instructions: "Always create and share the plan before execution."

Cowork may interpret a task and begin work with assumptions I did not intend. I review its plan first: which files it will read, what it expects to deliver, and how it will structure the work. I can then approve it, adjust the scope, or redirect it before it starts.

4. Set global instructions

These apply to every Cowork session regardless of folder. Go to Settings, then Cowork, then Global Instructions. Review these instructions when your working preferences change. Anthropic documents both global and folder instructions.

Mine:

My name is Prateek. I am Head of Product at Grab.
Check _Context.md first before starting any task.
Always share a plan before execution.
Be concise. Use simple English.
No em dashes. No corporate filler.
Use tables when comparing options.
Cite filenames when synthesizing across documents.

5. Layer folder instructions on top

Global instructions handle who you are and how you work. Folder instructions handle what you are working on right now.

When I am working on a market expansion project, the folder instructions might say: target market is Cambodia, payment architecture is X, key stakeholders are Y, deadline is H1. When I switch workstreams, I select the relevant folder and instructions. The project context stays contained instead of leaking across workstreams.

This layering is what makes the setup work across multiple projects without constant re-prompting. Global instructions stay the same. Folder instructions change per project.

6. Use parallel sub-agents for independent work

Cowork can spin up multiple sub-agents to work on separate pieces at the same time. It may do this on its own for complex tasks, or you can request it: "Spin up parallel sub-agents to research each of these five competitors separately."

Choose pieces that can be done independently. Researching five competitors is a useful candidate. If section 3 of a document depends on section 2, ask Cowork to resolve that dependency before drafting them.

Parallel work can use more of your usage allowance. Ask for a summary file at the end, such as what-changed.md, listing what each sub-agent did so you can review the combined result.

7. Automate recurring work with /schedule

Type /schedule in any Cowork task to set up jobs that run automatically on a daily or weekly cadence. This is where connectors make a real difference. I have Gmail, Google Calendar, Slack, and Google Drive connected, along with some internal work tools.

Here is what I have running right now:

Morning Brief (daily, 8:30 AM). Pulls unread emails, today's calendar, and overnight Slack activity. Produces a single-page summary with flagged action items.
End of Day Brief (daily, 6 PM). Summarizes what happened across email, Slack, and calendar. Highlights unresolved items. Drafts follow-ups where needed.
End of Week Brief (Friday, 5 PM). Synthesizes the full week. Decisions made, open items, suggested priorities for Monday.
Daily AI News Brief (daily, 8:30 AM). Searches the web for AI and product news, filters for relevance, delivers a short digest.

Anthropic now documents scheduled tasks as running remotely, including when your computer is asleep or Claude Desktop is closed. They use connected tools and files saved to your Claude account. Check the scheduled-task documentation when setting up a task that depends on local files.

Start with tips 1 through 4 to set up the context files and instructions. Once you have a setup that works for your tasks, consider whether sub-agents or scheduled work would help.