sf-ai-agentforce-persona
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates primarily through Markdown generation and does not contain any executable scripts, binaries, or malicious system commands.
- [SAFE]: File operations are restricted to the local
_local/generated/directory for saving design artifacts such as persona documents, scorecards, and sample dialogs. - [SAFE]: External references to GitHub repositories (jaganpro/sf-skills and cascadi/sf-ai-agentforce-persona) are documented for attribution and version control purposes and point to established development sources.
- [SAFE]: Data ingestion is limited to processing user-provided documents (e.g., brand guides) for signal extraction, which is an intended feature of the design workflow and does not involve exploitable logic.
- [SAFE]: The skill's metadata and YAML frontmatter are consistent with its stated purpose of persona architecture and do not contain deceptive instructions or metadata poisoning attempts.
Audit Metadata