linkedin-post-writer
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill encourages users to provide past content (10+ posts) for voice extraction and calibration. This establishes a surface for indirect prompt injection if the user-provided content contains malicious instructions.
- Ingestion points: User-supplied past posts provided in the chat context as requested in SKILL.md.
- Boundary markers: The instructions do not include specific delimiters or 'ignore-instructions' warnings for the data ingestion phase.
- Capability inventory: The skill does not request or use any system tools, shell commands, or network access.
- Sanitization: There is no specified method for filtering or sanitizing the input data provided by the user.
Audit Metadata