intercom-automation

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSNO_CODEPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to configure an external MCP server at https://rube.app/mcp, which is a third-party domain not associated with the official trusted vendors or the skill author's known infrastructure.\n- [NO_CODE]: The skill consists entirely of markdown instructions and metadata, with no accompanying scripts or executable files, reducing the risk of direct malicious code execution.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from external Intercom conversations and contacts.\n
  • Ingestion points: INTERCOM_GET_CONVERSATION, INTERCOM_LIST_CONVERSATIONS, and INTERCOM_SEARCH_CONTACTS in SKILL.md.\n
  • Boundary markers: No explicit instructions are provided to use delimiters or ignore instructions embedded within the Intercom data.\n
  • Capability inventory: The agent can take actions based on processed data, such as INTERCOM_REPLY_TO_CONVERSATION, INTERCOM_ASSIGN_CONVERSATION, and INTERCOM_CREATE_CONVERSATION.\n
  • Sanitization: The documentation advises users to sanitize HTML input to prevent rendering issues, but no automated sanitization is implemented in the skill itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 06:59 AM