meeting-prep

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were detected. The skill is instructional and its behavior aligns with the primary purpose of providing meeting research and sales intelligence.\n- [NO_CODE]: The skill consists entirely of markdown-based instructions and metadata. It does not include any executable scripts, binaries, or dependencies that could pose a direct execution risk.\n- [DATA_EXFILTRATION]: The skill manages professional contact information (emails, phone numbers). This is a core feature of the tool and is performed via the vendor's (Explorium) infrastructure. The instructions limit phone number retrieval to scenarios where the user explicitly intends to perform outreach.\n- [PROMPT_INJECTION]: An indirect prompt injection surface is present because the skill processes untrusted external data.\n
  • Ingestion points: LinkedIn posts, website changes, and business events (Step 3 and 4).\n
  • Boundary markers: Absent; the instructions do not specify delimiters to separate external content from the agent's instructions.\n
  • Capability inventory: The skill is limited to information synthesis; it possesses no file system write access, arbitrary command execution, or uncontrolled network capabilities.\n
  • Sanitization: The skill mitigates risks by instructing the agent to tag claims by source category and to flag stale data for verification.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 04:10 PM
Security Audit — agent-trust-hub — meeting-prep