AI Personal Assistant / Chief of Staff

How It Works

Tell it your role, tools, priorities and boundaries. This generates an EA agent blueprint the way an experienced automation strategist would build it - a real, usable deliverable, not a generic checklist. The output follows the standard: EA agent blueprint: role, tasks, tools/permissions, autonomy tiers, routines, guardrails. Replace every [[token]] with your specifics and it is ready to implement.

What to Provide

AI Personal Assistant / Chief of Staff

This is the finished deliverable.

1. Role definition and priorities/boundaries

Write the specific rule for your situation here: [[the concrete policy, threshold, or owner for role definition and priorities/boundaries]]. Base it on your role and adjust as real cases come in.

Signal of expertise this section should show: Defines explicit autonomy tiers (act vs ask), confirmation thresholds (e.g., never send externally without approval), and preference memory.

Mistake this guards against: Unbounded autonomy with no approval gates.

2. Task taxonomy

Cover each of these explicitly rather than leaving them implied: [[inbox]], [[calendar]], [[travel]], [[research]], [[drafting]]. Define [[the specific rule or default for inbox]] so nothing is left to guesswork.

Signal of expertise this section should show: respects calendar/email permission scopes.

Mistake this guards against: No privacy boundaries.

3. Tool/integration map and permissions

Cover each of these explicitly rather than leaving them implied: [[email]], [[calendar]], [[Slack]], [[CRM]]. Define [[the specific rule or default for email]] so nothing is left to guesswork.

Signal of expertise this section should show: Defines explicit autonomy tiers (act vs ask), confirmation thresholds (e.g., never send externally without approval), and preference memory.

Mistake this guards against: Vague task list without tool integrations.

4. Autonomy levels and approval gates

Write the specific rule for your situation here: [[the concrete policy, threshold, or owner for autonomy levels and approval gates]]. Base it on your role and adjust as real cases come in.

Signal of expertise this section should show: respects calendar/email permission scopes.

Mistake this guards against: Unbounded autonomy with no approval gates.

5. Standard operating routines

Cover each of these explicitly rather than leaving them implied: [[daily]], [[weekly]]. Define [[the specific rule or default for daily]] so nothing is left to guesswork.

Signal of expertise this section should show: Defines explicit autonomy tiers (act vs ask), confirmation thresholds (e.g., never send externally without approval), and preference memory.

Mistake this guards against: No privacy boundaries.

6. Escalation and confirmation rules

Write the specific rule for your situation here: [[the concrete policy, threshold, or owner for escalation and confirmation rules]]. Base it on your role and adjust as real cases come in.

Signal of expertise this section should show: respects calendar/email permission scopes.

Mistake this guards against: Vague task list without tool integrations.

7. Memory/preferences store

Write the specific rule for your situation here: [[the concrete policy, threshold, or owner for memory/preferences store]]. Base it on your role and adjust as real cases come in.

Signal of expertise this section should show: Defines explicit autonomy tiers (act vs ask), confirmation thresholds (e.g., never send externally without approval), and preference memory.

Mistake this guards against: Unbounded autonomy with no approval gates.

8. And privacy/data handling

Write the specific rule for your situation here: [[the concrete policy, threshold, or owner for and privacy/data handling]]. Base it on your role and adjust as real cases come in.

Signal of expertise this section should show: respects calendar/email permission scopes.

Mistake this guards against: No privacy boundaries.

Worked Examples

Example 1 - Series B startup CEO

Inputs: Protect deep work blocks, never double-book travel, no external sends without approval

Result: Autonomy-tiered EA agent handled 80% of scheduling without escalation while never sending anything externally unapproved.

Example 2 - VP of Sales managing a distributed team

Inputs: No meetings before 10am, no weekend emails, CRM + calendar tools

Result: Confirmation-gate rule caught a conflicting double-booked flight before it was ticketed.

Format Checklist

Common Mistakes to Avoid

• Unbounded autonomy with no approval gates.
• No privacy boundaries.
• Vague task list without tool integrations.

Next Steps After You Generate This

Week 1: pilot with a small internal group or a single channel/segment. Week 2: review real output against the format checklist below and fix the top 2-3 gaps. Weeks 3-4: expand scope and set a recurring review cadence so the workflow stays accurate as your data and process change.

This deliverable gives you a working starting point on day one - keep the [[tokens]] current as your process, tools, and volume change.