The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes financial data retrieved from external sources, which constitutes an attack surface for indirect prompt injection.
Ingestion points: Data is ingested through MCP tools (get_financial_statements, get_financial_ratios, get_growth_metrics, get_historical_valuation, get_treasury_rates, get_market_risk_premium, get_company_overview), web search results, and user-provided inputs as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or explicit boundary markers to isolate untrusted external content from the agent's instructions.
Capability inventory: The skill possesses the capability to execute Python scripts via Bash and perform file system operations (writing .xlsx files) using the openpyxl library.
Sanitization: There are no explicit instructions for sanitizing or validating the content of the retrieved financial data before it is used to generate formulas or scripts.
[DYNAMIC_EXECUTION]: The skill workflow involves the dynamic generation and execution of code to automate spreadsheet creation.
Evidence: SKILL.md instructs the agent to build the DCF model by saving a Python script (e.g., build_dcf.py) and running it via Bash. This is a routine operation for complex file generation using the openpyxl library.

dcf-model