skills/terminalskills/skills/intercom/Gen Agent Trust Hub

intercom

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The React component in SKILL.md dynamically loads the Intercom messenger script from https://widget.intercom.io. This is a well-known service and the official domain for Intercom's widget distribution.
  • [DATA_EXFILTRATION]: The skill includes server-side examples using fetch to send contact and message data to https://api.intercom.io. This behavior is expected for the skill's primary purpose of customer messaging and support.
  • [CREDENTIALS_UNSAFE]: No hardcoded credentials were found. The skill correctly demonstrates using environment variables (INTERCOM_ACCESS_TOKEN and NEXT_PUBLIC_INTERCOM_APP_ID) to manage sensitive integration tokens.
  • [SAFE]: Analysis of the instructions and code snippets revealed no signs of prompt injection, obfuscation, persistence mechanisms, or unauthorized privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 01:56 PM
Security Audit — agent-trust-hub — intercom