yy-sync-instructions-from-opencode

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a susceptibility to indirect prompt injection (Category 8) because it transfers natural language instructions from a configuration file into a primary guideline file.
  • Ingestion points: The skill reads the 'instructions' array from the local file '.opencode.json' (SKILL.md).
  • Boundary markers: There are no instructions to wrap the synchronized content in XML-like delimiters or to add warnings to the agent to ignore any embedded malicious instructions within the rules.
  • Capability inventory: The skill utilizes the agent's capability to read and write to the local file system (SKILL.md).
  • Sanitization: No sanitization, filtering, or validation is performed on the synchronized text to prevent the propagation of malicious prompt fragments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 02:56 AM
Security Audit — agent-trust-hub — yy-sync-instructions-from-opencode