
Agent Role Contracts
Catch AI-agent task conflicts before execution
9 followers
Catch AI-agent task conflicts before execution
9 followers
When an AI agent is assigned a task outside its declared scope—or asked to review its own work—things can go wrong before execution even starts. Agent Role Contracts checks role, authority, task scope, reviewer separation, and handoff declarations offline. Try the 3-minute PASS → FAIL → PASS demo with no API key or agent runtime. Free and MIT-licensed. It checks declarations, not runtime permissions.





Free
Launch Team / Built With

AWS Activate CreditsCredits that fuel your startup’s next milestone
Promoted




Quick update after launch:
One question we’re continuing to explore is where this kind of preflight check should live in a real agent stack.
Today, Agent Role Contracts is intentionally small and runtime-agnostic: it checks the declarations before handoff, while the runtime remains responsible for actually enforcing permissions.
I’m increasingly interested in the integration boundary:
Should this stay as an independent preflight primitive, or become a native layer inside orchestrators and agent runtimes?
If you’re building multi-agent systems or agent infrastructure, I’d be interested in how you’re handling this boundary today.
I’m looking for engineers working with coding agents who can share one recent example of a task going beyond its intended scope, a reviewer changing files, or a handoff losing important requirements. What happened, and what do you use to prevent it today?
I’m validating where ARC’s declaration checks add value alongside existing controls. It does not enforce runtime permissions. A short, non-confidential example by November 3 would help, including cases where your current tools already solve the problem. Thank you for sharing your experience.
Hey Tomohiro team, congrats on launching the agent role contract. As the agent will go more on autonomous tasks, we definitely need this. Congratulations on launching on Product Hunt.
@kartikrawatt Thanks, Kartik! Really appreciate your support.
As AI agents become more autonomous, we believe clearly defined roles, responsibilities, and handoffs will become increasingly important.
That's exactly what we're exploring with Agent Role Contracts — validating declared workflow contracts before execution.
Curious to hear your perspective: where do you see the biggest challenges as agents take on more autonomous work?