sales-analytics

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [SAFE]: No malicious code, exfiltration attempts, or persistence mechanisms were found. The skill operates within its stated purpose of sales reporting.\n- [PROMPT_INJECTION]: The skill ingests external data from the Attio CRM which is then processed by an AI node to generate summaries and recommendations. This constitutes an indirect prompt injection surface.\n
  • Ingestion points: Records from Attio People and Deals objects.\n
  • Boundary markers: The skill does not explicitly define delimiters for external data in the provided markdown.\n
  • Capability inventory: Network access via Attio REST API and Telegram API.\n
  • Sanitization: The logic relies on numerical extraction and structured JSON filters, but does not explicitly sanitize text-based CRM fields.\n- [EXTERNAL_DOWNLOADS]: The skill references the Attio API (api.attio.com) for data retrieval. These are well-known, legitimate technology services.\n- [CREDENTIALS_UNSAFE]: Authentication is handled through environment variable placeholders such as $ATTIO_API_KEY, which is the recommended method for preventing hardcoded secret exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 05:28 PM