Guardrails — CodeAgent Mobile Wiki
The CodeAgent Mobile Wiki

Every feature, explained — organized by the path you actually take: set up, launch a session, supervise the run, ship the PR.

Free

Guardrails

A per-session safety net: your agent asks before risky actions — you decide.

01 · AGENT
Hits a risky action
  • e.g. a force-push or reading a secret
02 · GUARDRAILPaused
The action pauses
  • The agent asks before acting
03 · YOUYou decide
Approve or deny
  • From your phone, wherever you are
04 · SESSION
Work continues
  • Approved actions proceed; denied ones stop

Guardrails give every session a safety net for risky agent actions. Each category — reading secrets, destructive commands, protected branches, and outward or irreversible actions like force-push, publish, or deploy — can be set to Deny, Confirm, or Off. Guardrails are on by default (Confirm): the agent asks first, and you approve or deny from your phone. They’re built to catch accidental risky actions — not a hard security boundary.

On by default

Every category starts on Confirm, so the agent asks before anything risky — no setup needed.

Per-category, per-session

Tune each category for each session: Deny blocks the action outright, Confirm asks you first, Off trusts the agent.

Decide from anywhere

Confirmation requests reach your phone — approve or deny without being at your desk.

A safety net, honestly

Designed to catch accidental destructive actions. It’s not a hard security boundary — use environment-level restrictions when you need hard guarantees.

  1. 1
    Open the session’s ⋯ menu → Guardrails

    Guardrails are configured per session, from the session menu.

  2. 2
    Set each category

    Choose Deny, Confirm, or Off for secrets, destructive commands, protected branches, and outward actions.

  3. 3
    Save

    The settings apply to the session right away — the agent asks (or is blocked) from the next risky action on.

FAQ

Related

Ready to try it?