System Telemetry
Scanning Statistics
Real-time transparency into the state of AI agent skill security.
Looking for the snapshot research report? Read the Intelligence Report →
Resume your workflow
We saved your last report and scan target so you can jump from telemetry back into the active trust workflow.
10,633
Skills Scanned (Live)
94
Avg Trust Score
46,924
Total Findings
9,188
Certified
Badge Distribution
CERTIFIED
9188 (86%)
CONDITIONAL
873 (8%)
SUSPICIOUS
287 (3%)
REJECTED
285 (3%)
Scan Sources
ClawHub Registry
6,198 (58%)
Public API● api
3,063 (29%)
Website Submit Page● website
1,369 (13%)
Admin
3 (0%)
1,369 skills submitted from the website
3,063 skills submitted via API clients
Most Common Findings
lowUnknown external reference21322×
lowNo explicit safety boundaries8907×
infoSafety boundaries defined1625×
infoOutput constraints defined1434×
infoError handling instructions present1426×
lowMissing or insufficient description1167×
highCapability contract mismatch: inferred command execution is not declared878×
mediumFinancial/payment actions detected751×
mediumCapability contract mismatch: inferred network access is not declared726×
mediumSystem modification detected (inside code block)695×