The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill integrates untrusted external data from the internet to drive its core logic and decision-making processes.
Ingestion points: Phase 2, Step 3 instructions specify using WebSearch to collect related news and trends from the past two weeks, which are then used to build scenarios and identify stock impacts.
Boundary markers: None. There are no instructions provided to the agent to treat external content as untrusted or to wrap it in delimiters to prevent the execution of embedded instructions.
Capability inventory: The skill utilizes the WebSearch tool and get_ltp market data tool, and it performs file-write operations to the reports/ directory.
Sanitization: There is no evidence of content filtering, escaping, or validation of the data retrieved via web searches before it is interpolated into the analysis workflow.

scenario-analyzer