meeting-prep
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified in the skill file.
- [PROMPT_INJECTION]: The instructions are task-specific and do not contain patterns aimed at bypassing AI safety guardrails or overriding system prompts.
- [DATA_EXFILTRATION]: The skill does not perform any network requests or file system operations. It operates entirely on context provided to the agent and does not handle sensitive credentials.
- [REMOTE_CODE_EXECUTION]: There are no indicators of external script downloads or package installations. The skill is purely instruction-based.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data such as email reply threads and prospect profiles (SKILL.md).
- Ingestion points: Reads from 'G6 handoff context' (reply threads) and 'prospect-profiler card'.
- Boundary markers: None explicitly defined for input data, though output instructions require tagging non-grounded data as '[inferred]'.
- Capability inventory: The skill is limited to text generation and lacks shell execution, network access, or file-writing capabilities.
- Sanitization: No explicit sanitization of external content is defined in the instructions.
- Assessment: While an ingestion surface exists, the lack of dangerous capabilities prevents exploitation via this vector.
Audit Metadata