think-framework-advisor
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and data-driven, consisting of Markdown files and a JSON reference list. It does not utilize any tools, execute shell commands, or perform network operations.
- [PROMPT_INJECTION]: The skill contains explicit instructions to handle potential prompt injection attempts within user input. It specifies that any directives found within user-provided content (e.g., 'ignore the above') should be treated as material to analyze as part of the situation, rather than instructions for the agent to follow.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted user data (decisions, problems, notes), the risk is mitigated by a lack of executable capabilities. The agent is instructed to use a 'source ledger' to ground all claims in exact quotes from the input, which provides a clear audit trail and prevents the fabrication of information.
- [SAFE]: No hardcoded credentials, sensitive file paths, or obfuscated content were detected within the skill files.
Audit Metadata