jquants-cli-usage
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external market data, financial statements, and corporate disclosures from the J-Quants API. This external content represents a potential indirect prompt injection surface as it is ingested into the agent's context.\n
- Ingestion points: Multiple commands (e.g.,
jquants eq,jquants fins,jquants td) fetch data from the J-Quants API (SKILL.md).\n - Boundary markers: No specific boundary markers or instructions to ignore embedded prompts are provided in the command templates.\n
- Capability inventory: The skill allows writing fetched data to the local file system via the
--saveflag.\n - Sanitization: Data is processed as CSV/JSON/Table output from the CLI tool without additional sanitization instructions.\n- [SAFE]: The skill provides legitimate guidance on using the
jquantsCLI. It correctly identifies the local credential storage path (~/.config/jquants/credentials.json) created by the tool's login process and uses standard environment variables for API keys. No suspicious network operations or privilege escalation attempts were detected.
Audit Metadata