Policy-as-code for agent execution

Agent policies should be versioned, testable, reviewable, and applied by the runtime.

Policy hidden in prompts is hard to audit. Policy-as-code gives agent execution clear, testable constraints.

AstraRunner treats this as an execution-platform concern: the run needs durable state, traceable decisions, cost visibility, and enough structure for teams to review what happened after the agent finishes.

Patterns that work

• Represent allowed tools, budgets, approvals, and data scopes as policy.
• Version policies with release notes.
• Test policy decisions against examples.
• Log policy decisions in traces.

Implementation checklist

• Separate policy from prompt text.
• Add policy tests to CI.
• Expose policy version per run.
• Require review for policy changes.

Watch outs

• Prompt-only policy drifts silently.
• Policy engines need context quality.
• Too-rigid policy can block legitimate work without escalation paths.

Related reading

• AstraMemory on compliance and right to forget

How this fits the Astra stack

AstraRunner owns the operational path for agent work: orchestration, scheduling, approval gates, traces, costs, and role-agent handoffs. AstraMemory owns durable context and retrieval. AstraGenie turns those capabilities into product workflows for teams that want automation without rebuilding the platform layer.