research-synthesis-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured methodology for research synthesis, including source evaluation, note-taking templates, and thematic analysis frameworks.
- [DYNAMIC_EXECUTION]: The file
references/synthesis-methods.mdcontains Python code snippets for calculating effect sizes (Cohen's d) and performing fixed-effects meta-analysis. These snippets use standard libraries (numpy,scipy) and perform benign statistical calculations without any network or file system access. - [INDIRECT_PROMPT_INJECTION]: While the skill is designed to ingest and process external research data, it includes robust 'Source Quality Assessment' and 'Bias Detection' rubrics to mitigate risks associated with untrusted information. It lacks dangerous capabilities like arbitrary command execution or network exfiltration, making the surface area for this attack vector negligible.
Audit Metadata