composing-linkedin-stories
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): The skill instructs the agent to process external LinkedIn profile links and post examples, which are untrusted data sources. This creates a surface for indirect prompt injection.
- Ingestion points: Discovery Question 4 (LinkedIn profile link and example posts).
- Boundary markers: Absent; there are no instructions to delimit the external data or to disregard any embedded commands within it.
- Capability inventory: Professional experience transformation, content analysis, and narrative generation.
- Sanitization: Absent; no escaping or validation of the fetched content is specified.
- NO_CODE (SAFE): The skill consists entirely of markdown-based prompt instructions and does not include any Python scripts, Node.js packages, or shell commands, effectively eliminating the risk of direct remote code execution or system compromise.
Audit Metadata