isr-topic-selection
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely instructional and contains no code, scripts, or binaries. No evidence of obfuscation, malicious command execution, or unauthorized network access was found.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for analyzing user-provided research phenomena and datasets, creating a surface for indirect prompt injection. However, the lack of dangerous capabilities mitigates this risk. 1. Ingestion points: User-provided research phenomena, datasets, and models described in the trigger section. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present. 3. Capability inventory: No tools, subprocess calls, network operations, or file-write capabilities are requested or used. 4. Sanitization: No sanitization or validation of the input phenomenon is described.
- [NO_CODE]: The skill consists exclusively of markdown text providing guidance and does not include any scripts or automation logic.
Audit Metadata