linkfox-shopee-store-media-space

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/_shopee_media_space_common.py utilizes subprocess.run to execute a local Python dependency check script scripts/check_auth_dependency.py at runtime.
  • [DATA_EXPOSURE]: The script scripts/check_auth_dependency.py probes multiple application-specific directories in the user's home directory (such as .claude/skills, .cursor/skills, and .linkfox/skills) to verify the presence of required dependency skills.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Shopee API responses and provides summaries to the agent, creating a surface for indirect instructions.
  • Ingestion points: Shopee API response bodies processed in scripts/_shopee_media_space_common.py and summarized by the tool.
  • Boundary markers: Absent; there are no explicit instructions or delimiters in the prompt to warn the agent against embedded instructions in the API data.
  • Capability inventory: The skill has the ability to write JSON files to the local workspace, make network requests to vendor-owned domains, and execute internal scripts.
  • Sanitization: Content is handled as JSON objects, which prevents structural injection but does not filter or sanitize natural language content for malicious instructions.
  • [PERSISTENCE_MECHANISMS]: The documentation in references/onboarding.md provides instructions for the user to manually append environment variable exports to shell profile files like .zshrc and .bashrc to maintain persistent authentication tokens across sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 07:03 AM
Security Audit — agent-trust-hub — linkfox-shopee-store-media-space