longbridge-portfolio-rebalance
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions utilize the
longbridgeCLI tool to retrieve account holdings and portfolio metadata. This execution is limited to data retrieval for analysis purposes and is consistent with the vendor's provided tooling. - [EXTERNAL_DOWNLOADS]: The skill references an official vendor MCP endpoint (
https://openapi.longbridge.com/mcp) for data access when the local CLI is missing. This is a trusted infrastructure component belonging to the skill author. - [DATA_EXFILTRATION]: The skill accesses sensitive financial data (portfolio positions and total value) to perform calculations. This data remains within the agent's context for recommendation generation and is not transmitted to unauthorized external domains.
- [PROMPT_INJECTION]: A potential surface for indirect prompt injection exists via user-provided target weights. However, the risk is negligible as the skill is strictly read-only, has no capability to execute trades, and its logic is confined to deterministic mathematical drift analysis.
Audit Metadata