social-kit

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates the execution of other agent skills (create-x-content, create-linkedin-content, and goose-graphics) based on user-provided arguments derived from the brief.\n- [PROMPT_INJECTION]: The skill ingests untrusted user text through the --brief parameter, which is interpolated into downstream skill calls. This represents an indirect prompt injection surface.\n
  • Ingestion points: SKILL.md (Steps 1, 3, 4, 7).\n
  • Boundary markers: Arguments are enclosed in double quotes within the command templates.\n
  • Capability inventory: Invokes other skills and creates content directories via the agent.\n
  • Sanitization: Relies on standard shell-style quoting for argument separation.\n- [SAFE]: The skill accesses local configuration files in ~/.goose-skills/ to retrieve user preferences and voice guides. These operations are consistent with the vendor's own configuration management and do not involve data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 02:14 PM
Security Audit — agent-trust-hub — social-kit