lead-researcher
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a complex research orchestrator that uses best practices for information gathering and synthesis, including built-in verification phases such as 'stream per-finding verify' and 'CRITIC refute'.
- [EXTERNAL_DOWNLOADS]: Research data is fetched from well-known and reputable technical forums, academic registries, and official documentation sites across multiple languages. These sources (e.g., GitHub, arXiv, Lobste.rs, V2EX, Habr) are essential for the skill's stated functionality and are documented as target sites for research.
- [COMMAND_EXECUTION]: The skill employs standard development tools like the GitHub CLI (gh) and performs local file operations to manage research state and results. These operations are standard and appropriately scoped to the research tasks.
- [DATA_EXFILTRATION]: Network operations are limited to research queries on public and specialized technical domains. No evidence of unauthorized harvesting or transmission of sensitive local data or credentials was found.
- [PROMPT_INJECTION]: The skill's instructions are designed to coordinate a complex research workflow using an 'Adaptive Iterate Loop'. They do not contain patterns intended to override AI safety guidelines or bypass instructions.
Audit Metadata