The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uv run to execute various analysis modules (e.g., src.analysis.risk_metrics_cli, src.strategies.backtester_cli). This is the primary mechanism for generating quantitative metrics and backtesting results.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including the financial-datasets MCP and web searches for financial filings.
Ingestion points: External financial data and web search content referenced in SKILL.md.
Boundary markers: Includes instructions to define a plan and validate assumptions against compliance policies.
Capability inventory: Shell command execution via uv run for various statistical and optimization scripts.
Sanitization: No explicit validation or filtering logic for external data inputs is documented within the skill instructions.

fin-guru-quant-analysis