linkedin-post
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill content is restricted to natural language instructions for text generation, formatting, and engagement strategies for LinkedIn.
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions to ingest data from the
/knowledge/directory and specific profile files in/context/. While this represents a surface for indirect prompt injection, it is central to the skill's purpose of personalizing content. - Ingestion points:
/knowledge/directory and files/context/voice-dna.json,/context/icp.json, and/context/business-profile.json. - Boundary markers: None defined in the skill instructions.
- Capability inventory: The skill has no capabilities to execute shell commands, perform network operations, or write to the filesystem.
- Sanitization: No sanitization or filtering of input data is specified.
- Risk Assessment: The absence of actionable tools or system-level access ensures that any instructions embedded in the knowledge files cannot result in privilege escalation or data exfiltration.
Audit Metadata