DiagnosticsFree
TriageRadCheck
Start here. Triage diagnoses your stack. RadCheck scores it 0–100.
Run Triage free →
Field Supply Division
Mission Management Software for OpenClaw
Stabilize. Coordinate. Results.
Make it work (for you)
Control your stack. Lead your agents.
Pick your starting point.
Full pricing →Start free. Add protection as you need it. No lock-in.
Runtime$5 / mo
SentinelWatchdog
Sentinel watches for silent failures. Watchdog catches loops, stalls, and double-runs.
Add runtime protection →Incident Response$19 / mo
Agent911RecallLazarus
When something breaks at 2am, you open Agent911. Recall gives you manual intervention. Lazarus confirms your backups actually restore. Everything you need to recover, in one place.
See Incident Response →Operator Bundle$29 / mo
SentinelInfraWatchWatchdogLazarusAgent911Recall
The full resilience layer, fully wired. Sentinel detects. InfraWatch watches config drift. Watchdog catches dead processes. Lazarus confirms recovery. Agent911 executes. You get a report.
See the full layer →Works when OpenClaw doesn't.
Reads directly from the filesystem. No gateway required. Run it when the system itself is the suspect — or when protection_state AT_RISK and nothing has broken yet.
STATUSHEALTHY
reliability_score83
protection_stateAT_RISK
sessions17 active, 0 orphan
runtime alerts72,833 captured
Read-only. Nothing changes. GPG-verified install.
What You Can See
Most operators don't know something is wrong until it's already wrong. ACME surfaces the signals that tell you earlier.
- Agent topologyWhich agents are running, where, and how they relate to each other.
- Agent activity rateHow active each agent is — and when activity patterns look wrong.
- Runtime alertsSignals that something changed in the runtime before it becomes a failure.
- Reliability scoreA 0–100 score that tells you how healthy your stack is right now.
- Protection eventsWhen Sentinel or Watchdog acted — what it caught and when.
When Something Breaks
ACME tells you what to do — in order, with evidence, without guessing.
ORP
Recovery Protocol
Sequences recovery correctly every time. Evidence first, then diagnosis, then action.
Agent911
Recovery Cockpit
Your incident command surface at 2am. Health signals, anomaly classification, and structured guidance on what to do next.
Lazarus
Backup Readiness
Verifies your backups actually work before an incident forces the test. Lazarus finds out first.
Operator Stories
Memory Drift Recovered
Agent system recovered from memory drift before cascading impact.
MTTR: 11m
Gateway Stall Prevented
Gateway stall detected and classified before downstream failure.
Incident scope: constrained
Reliability Lift
Reliability score improved after deterministic recovery workflow.
46 → 74 in 24h
Where we're going
Five patents pending, all about one thing: giving operators like you real control over multiple agents
Not just keeping agents alive — leading them, coordinating them, measuring what they produce. The floor is reliability. ACME gets you there, and beyond.
Stick with us
Stay in the loop
Get updates on new releases
Early access, release notes, and operator field updates. No noise.