Fact Checker
Fact Checker · owned by Admin
Checks everything the other job agents write for her before she sees it: each claim about her against her FACTS, each claim about a company against its cited source, every date against its weekday, and her voice rules. Anything it can’t verify is flagged, never passed as fact. It only flags: it never edits, sends or submits.
Autonomy
The trust dial: how freely the agent acts on things it's already allowed to do. Even at full autonomy, it can never reach beyond its permissions.
Acts on its own — within scope + guardrails
Scope
What the agent may touch. Anything not explicitly allowed here is off-limits by default — no exceptions.
| Read | Sync | Flush | Restart | Roll back | Rotate | Invoke | Draft | Notify | Move to Trash | Submit | Send | Configure | Purchase | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Service | ||||||||||||||
| Integration | ||||||||||||||
| Deployment | ||||||||||||||
| Data store |
Default deny — access exists only where explicitly granted.
Guardrails
Safety rules on top of permissions — the strictest rule wins. A guardrail can send even a fully-autonomous agent to a human for approval.
Ownership
Every agent answers to a responsible human. If that person leaves, the agent pauses itself — nothing keeps running with nobody accountable.
Activity
What this agent has attempted, each decided live by the engine. Run its next scripted action, or type one in plain language.
No attempts yet — run an action and watch Warden decide.