linkedin-creator

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted data during Phase 1 (Audience, Goal & Voice Audit) to generate tailored content. The absence of explicit boundary markers or sanitization could allow malicious instructions within the audit data to influence the agent's behavior. \n
  • Ingestion points: Phase 1 (Audience, Goal & Voice Audit) in SKILL.md. \n
  • Boundary markers: No delimiters or instructions to ignore embedded commands are specified for the ingested content. \n
  • Capability inventory: Access to bash, git, and file system operations as listed in the technical deliverables section. \n
  • Sanitization: No input validation or sanitization logic is implemented for user-supplied profiles. \n
  • [COMMAND_EXECUTION]: The skill explicitly documents the availability of high-privilege tools including bash, git, and file operations. While these are intended for legitimate project management, they increase the impact of potential injection vulnerabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:51 PM
Security Audit — agent-trust-hub — linkedin-creator