model-currency-checker

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute pnpm starter:audit and suggests using search tools like grep to identify model IDs within the source code.
  • [DATA_EXFILTRATION]: The instructions require the agent to scan sensitive files including .env and .env.example. While these files often contain secrets, the access is scoped to searching for model version identifiers during a local audit and does not involve network exfiltration.
  • [PROMPT_INJECTION]: An indirect prompt injection surface is present as the skill processes untrusted data from multiple file types. \n
  • Ingestion points: Codebase files including *.ts, *.js, *.py, *.json, *.yaml, *.yml, *.env, and .env.example. \n
  • Boundary markers: None identified. \n
  • Capability inventory: File reading, report generation (reports/model-currency-report.md), and source code modification (remediation). \n
  • Sanitization: No sanitization or verification of file content is described before the data is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 11:48 AM