longbridge-fundamentals

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data such as financial news, company announcements, and regulatory filings fetched via the longbridge news and longbridge-corporate commands. This data represents an untrusted ingestion surface where malicious actors could attempt to influence the agent's output through embedded instructions in public financial records or news articles. However, this is an inherent risk for financial analysis tools and the skill operates within a read-only risk level.
  • Ingestion points: longbridge news, longbridge company, and longbridge filings subcommands (referenced in references/corporate-events.md and references/main-business-analysis.md).
  • Boundary markers: The instructions do not specify explicit delimiters for external content.
  • Capability inventory: Shell execution of the longbridge CLI and Python sandbox for calculations.
  • Sanitization: Not explicitly mentioned in the instructions.
  • [COMMAND_EXECUTION]: The skill makes extensive use of a proprietary CLI tool (longbridge) to fetch financial data. It also includes instructions to run longbridge update to maintain the tool's functionality. Since these operations are directed at the vendor's own infrastructure and tools, they are consistent with the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 12:25 PM