intercom-analysis

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted customer conversation data, creating a surface for indirect prompt injection.\n
  • Ingestion points: The agent fetches conversation message bodies, internal notes, and contact profiles using the tools 'get_conversation' and 'get_contact' as documented in 'references/mcp-tools.md'.\n
  • Boundary markers: There are no instructions defining delimiters or warnings to ignore potential instructions embedded within the conversation data fetched from Intercom.\n
  • Capability inventory: The agent's capabilities are restricted to searching and fetching Intercom-specific data; no tools for arbitrary shell execution, local file writing, or non-whitelisted network access are present.\n
  • Sanitization: The instructions do not mention any sanitization or validation of the retrieved text content before it is processed for pattern analysis or summarization.\n- [NO_CODE]: The skill package consists entirely of instructional markdown and reference documentation, with no executable scripts, binaries, or configuration files that could hide malicious logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 09:30 PM