meeting-analyzer
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, suspicious network activity, or unauthorized data access behaviors were detected. The skill is designed to perform text-based analysis of local files as part of its primary function.\n- [NO_CODE]: The skill is composed entirely of natural language instructions and does not include any executable scripts, binary files, or external package dependencies.\n- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes untrusted meeting transcripts provided by the user. However, the skill does not have access to high-risk tools (e.g., shell execution or network requests) that could be used to exploit such an injection.\n
- Ingestion points: Scanning and reading local transcript files in .txt, .md, .vtt, .srt, .docx, and .json formats from user directories.\n
- Boundary markers: None specified. The instructions do not establish clear delimiters to separate transcript data from the agent's logic.\n
- Capability inventory: Limited to reading local files, normalizing text, and generating analytical reports.\n
- Sanitization: No sanitization or filtering logic for the content of processed transcripts is defined in the instructions.
Audit Metadata