Step 01
Define a Project and Assign an Owner
Create a 'Project' within your AI platform to act as a governance container for a specific team or workflow. Assign a Directly Responsible Individual (DRI) as the owner.
AdvancedEngineering
How to Implement AI Governance with Context-Aware Project Controls
Create context-aware guardrails for your enterprise AI agent using a 'Project' governance layer. As a project owner, define rules, scope tools, and require human-in-the-loop approval for sensitive actions, ensuring safe and effective AI use.
Try this with your AI assistant
Download the skill for this workflow. Version 1.0.1.
Source episodeHow Stripe Built Kai: Data Dashboards, Reusable Skills, and Enterprise AI Governance
with Claire Vo
Before you start
Tools used
- ClaudeVisitClaude
Anthropic AI assistant
The workflow
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
4 steps
Step 02
Configure Project-Specific Settings
In the project settings, configure defaults to manage cost and performance. This can include specifying a default LLM (e.g., a cheaper model like Claude Code) or scoping the available AI skills to only those relevant to the project.
Step 03
Set Human-in-the-Loop Tool Policies
Define policies for sensitive tools. For a specific project, you can require explicit human approval before the AI agent is allowed to use a tool like Google Calendar or access a private data source.
This creates 'appropriate friction,' shifting the security burden from the end-user to the platform, which enforces context-specific policies.
Step 04
Test the Governance Policy
Within the project's context, prompt the AI to perform an action that triggers the policy to ensure the human-in-the-loop confirmation dialog appears as expected.
Example prompt
Create a calendar invite for me and Huang tomorrow at 11 AM Pacific
Build your next product with ChatPRD
Turn an idea into a PRD, user stories, and a plan.
Related workflows
intermediateEngineeringMeta MuseClaude Code
How to Set Up Meta's Muse Wonder for Headless AI Testing
A step-by-step guide to signing up for Meta's Muse API, installing the Muse Code command-line tool, and running it headlessly with an external agent like Claude Code for easier testing and benchmarking.
Open
advancedOperationsMeta Muse
How to Test an AI Agent for Complex Customer Support Ticket Resolution
Benchmark an agentic AI model's ability to resolve complex customer billing issues. This workflow tests the AI's ability to investigate tickets, handle complex logic, execute changes in an admin app, and communicate with the customer.
Open
intermediateProductClaudeGoogle Drive+1
Generate New Video Ideas by Analyzing Your TikTok Comments with Claude
Use your existing TikTok video library and their associated comment threads to generate highly relevant new content ideas and scripts. This workflow pairs audience feedback directly with the specific video performance that prompted it, leading to targeted and engaging new topics.
Open
Start shipping better products.
Join 250,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.
Free to start · 14-day Pro trial · SOC 2 Type II
