opencode-skill-usage

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill retrieves usage statistics from the local OpenCode database using standard command-line tools.
  • [SAFE]: Database queries are executed in read-only mode using the -readonly flag, preventing any accidental or malicious modification of data.
  • [SAFE]: The shell script implements rigorous validation on the user-provided limit parameter, allowing only positive integers via regular expression matching to prevent injection vulnerabilities.
  • [DATA_EXPOSURE]: The skill reads from ~/.local/share/opencode/opencode.db to extract skill names and usage counts. This access is limited to the skill's stated purpose of providing usage statistics and does not exfiltrate data externally.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 11:45 AM
Security Audit — agent-trust-hub — opencode-skill-usage