The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the web using the web_search tool to analyze market conditions.
Ingestion points: Data is collected from external financial news and official sources via web_search as described in SKILL.md.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: The skill only performs read-only analysis and formatting via scripts/market_utils.py.
Sanitization: Standard agent handling of search results is assumed.
Assessment: This represents a standard attack surface for analysis-oriented skills and does not pose a direct threat as the skill lacks dangerous write or execution capabilities.
[COMMAND_EXECUTION]: The skill instructions in SKILL.md reference the execution of a local Python script scripts/market_utils.py.
Evidence: python scripts/market_utils.py.
Assessment: The script is provided within the skill package and contains only benign utility functions for formatting dates, calculating trading days, and categorizing market volatility using the standard datetime library. It does not perform any dangerous operations, network requests, or file system modifications.

market-environment-analysis