kill-switch
Halt the desk through risk.kill_switch set or flatten_request, read state through risk.halt_status, and clear only through the human CLI path.
SKILL.mdskills/kill-switch/SKILL.md
GitHubKill Switch
L1 - Summary
The kill switch halts the desk. RISK sets it with the risk.kill_switch tool (set or flatten_request only); CHIEF watches it with risk.halt_status. The switch can be set from the environment at start, or from the var/desk/KILLSWITCH file (the one latch path; an alert KILL from an allowlisted Telegram chat writes this same file). Clearing is CLI-only with a human-signed record; no skill and no seat clears the switch.
threat seen -> RISK set (or file/env/alert trips it) -> halt: no new tickets
flatten_request -> in-flight orders settle, open positions close with reason
clear -> pumpgrok killswitch clear --by <human> --reason <text> only
L2 - Method
- S1 - Read the state first: call
risk.halt_statusto learn whether the desk is halted and which source tripped it (tool action, file, environment, or alert channel). CHIEF polls this every cycle. - S2 - Set on danger: RISK calls
risk.kill_switchwithaction=seton a confirmed threat (suspected rug in inventory, daily-loss breach, integrity failure). The action enum accepts onlysetandflatten_request;clearis a validation error. - S3 - Flatten when exposed: with open positions, RISK uses
action=flatten_requestso the supervisor waits on the adapter barrier and closes every position with the kill reason instead of abandoning it. - S4 - Honor every source: a
var/desk/KILLSWITCHfile present at start, the environment flag, or an alert KILL from an allowlisted chat all halt before the first ticket. Removing the file does not clear a tool-set switch. - S5 - Clear human-only:
pumpgrok killswitch clear --by <human> --reason <text>with an approval record. Seats never clear; CHIEF requests, the human clears.
Failure modes
| Symptom | Cause | Action |
|---|---|---|
| Switch set but tickets continue | A seat bypassing the halt read | Halt the run; every ticket path reads halt status first |
| Clear attempted by a seat | Misunderstood authority | Refuse; only the human CLI path with a signed reason clears |
| Order stuck ACCEPTED at halt | In-flight order during the trip | Reject at CONFIRMING with reason kill_switch; never fill into a halt |
| Positions left open after flatten | Barrier or adapter failure | Escalate to RISK; keep polling until every position reports closed |
References
- Tools:
risk.kill_switch(RISK-only grant;setplusflatten_request),risk.halt_status(every halted-seat read) - CLI:
pumpgrok killswitch clear --by <human> --reason <text>(sole clear path) - Companion skills:
live-gate-procedure(preconditions),desk-incident-response(halt playbooks)