Policy checks for AI coding agents: decides allow, defer, ask or deny for every tool call from Claude Code, Codex and Cursor, configures Claude Code's and Codex's own OS sandboxes from the same policy, and keeps a local audit log.
//! Other systems (Windows): the parent processes are not read yet, so only
//! the terminal check applies to the human-only commands.
usestd::io;pub(super)fnagent_ancestor()->io::Result<Option<String>>{// Pid 0 ends the walk at once: no parent is known here.
super::walk(0,|_|{Err(io::Error::other("parent processes are not read on this system",))})}