skills/pubnub/skills/pubnub-presence/Gen Agent Trust Hub

pubnub-presence

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides standard documentation and code samples for implementing PubNub Presence features, including online/offline indicators and occupancy tracking.
  • [DATA_EXFILTRATION]: No sensitive data access or unauthorized network operations were found. The skill correctly uses placeholders (e.g., 'pub-c-...') for API keys and sensitive configuration.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes handling user-controllable metadata ('state') via presence events.
  • Ingestion points: Presence event objects processed in the listener (references/presence-events.md).
  • Boundary markers: None (data is handled within JavaScript application logic).
  • Capability inventory: Console logging and UI updates (e.g., 'addUserToList'). No high-risk capabilities like shell execution are involved.
  • Sanitization: Uses standard 'JSON.stringify' for logging state objects.
  • [COMMAND_EXECUTION]: The skill contains no shell commands, system modifications, or dynamic execution of untrusted scripts.
  • [PRIVILEGE_ESCALATION]: The skill promotes security best practices by detailing how to use PubNub Access Manager to grant least-privilege permissions to presence-specific channels ('-pnpres').
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 12:07 PM