skills/openclaw/skills/messenger/Gen Agent Trust Hub

messenger

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONNO_CODE
Full Analysis
  • Prompt Injection (LOW): The skill establishes a pipeline for receiving and processing data from Facebook Messenger webhooks, which constitutes an indirect prompt injection surface.
  • Ingestion points: references/webhook-event-map.md (inbound messages and postbacks).
  • Boundary markers: Absent. The skill does not provide delimiters or instructions for the agent to distinguish between system instructions and untrusted user input.
  • Capability inventory: references/messaging.md (Send API) allows the agent to perform network-based messaging actions.
  • Sanitization: Absent. The provided templates do not include logic for sanitizing or escaping user-generated content.
  • Data Exfiltration (LOW): The skill specifies network communication with an external API domain not included in the trusted whitelist.
  • Evidence: references/messenger-api-overview.md points to https://graph.facebook.com for API operations.
  • No Code (SAFE): No executable scripts, binaries, or automated installation packages are included. The skill consists entirely of documentation and configuration templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 01:51 AM