social-selling-planner
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown documentation and configuration. No executable scripts (Python, JavaScript, shell) or binaries are included, which significantly reduces the technical attack surface.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data such as social signals, scraped text, and reviews. However, it explicitly mitigates potential prompt injection risks by instructing the agent to treat these sources as untrusted input and to ignore any embedded instructions within that data (documented in
SKILL.mdandreferences/skill-contract.md). - [COMMAND_EXECUTION]: There are no commands or dynamic context injection patterns (e.g., backtick-enclosed shell commands) detected in the skill files.
- [DATA_EXFILTRATION]: The skill does not contain hardcoded credentials or instructions to exfiltrate sensitive data. It explicitly requires user authorization for any external side effects or account mutations.
Audit Metadata