persona-researcher
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of markdown instructions and YAML metadata. No scripts, binaries, or executable code files are present in the package.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading external research documents, which is an inherent ingestion surface for indirect prompt injection. However, given this is the core intended purpose of the persona, it is classified as a standard operational risk.
- Ingestion points: Reads research papers and notes via
gws-drive(SKILL.md). - Boundary markers: Not specified.
- Capability inventory: Includes file modification (
gws-docs,gws-sheets) and email communication (gws-gmail) (SKILL.md). - Sanitization: None defined.
- [SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or other malicious patterns was found. The skill operates within the expected scope of the provided Google Workspace tools.
Audit Metadata