fii-dii-flow-tracker
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves institutional flow data by querying external websites and depositories such as NSDL, NSE, and MoneyControl.
- Ingestion points: SKILL.md (Steps 1, 2, and 7) utilizes
WebSearchandWebFetchto ingest data from third-party financial news and official depository sites. - Boundary markers: The instructions lack explicit delimiters or instructions to ignore potential commands embedded within the retrieved external content.
- Capability inventory: The skill is configured to use broker-specific MCP tools (Groww and Zerodha Kite) which include sensitive capabilities such as
get_equity_portfolio_holdingsandget_holdings. - Sanitization: There is no documented process for sanitizing or filtering the fetched content before it is processed by the agent, creating a surface where malicious instructions on a web page could theoretically influence the agent's actions.
Audit Metadata