Overview
During an active Pi native run, RStack intercepts everytool_call event. Certain destructive commands and sensitive file paths are blocked by default — they require explicit human approval before they can execute.
This prevents AI agents from accidentally shipping code, destroying data, or leaking secrets.
Blocked shell commands
Repository operations
Repository operations
Filesystem destruction
Filesystem destruction
Package publishing
Package publishing
Infrastructure changes
Infrastructure changes
Database mutations
Database mutations
Blocked write paths
RStack also blocks writes to files that match these patterns — to prevent secret exfiltration or credential overwriting:Approving a protected action
Via sdlc_approve
approvals.json. The agent can then execute the blocked action once.
For release-level actions (deploys, publishes), approve the release artifact:
Via environment variable
To bypass all protections for a session (use with caution):Protection scope
Protection via
tool_call hooks is a Pi-native feature. For other adapters, the governance model relies on the agent following the operating standard in agents/core/orchestrator.md — which instructs it to ask before destructive actions.
