The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from SEC filings and web searches which could theoretically contain malicious instructions designed to influence the agent's analysis or output.
Ingestion points: Financial data is retrieved from SEC EDGAR filings, analyst reports, and general web searches as described in the workflow section of SKILL.md.
Boundary markers: The instructions do not provide specific delimiters or warnings to the agent to ignore potentially adversarial instructions embedded within the ingested financial documents.
Capability inventory: The skill executes Python scripts (validate_dcf.py, recalc.py) and writes to the local filesystem using openpyxl.
Sanitization: There is no explicit requirement for the agent to sanitize or filter the text content retrieved from external sources before it is used in the model summaries.
[COMMAND_EXECUTION]: The skill requires the execution of local Python scripts to validate the integrity of the generated Excel files.
Evidence: SKILL.md instructs the agent to run python recalc.py model.xlsx 30 and python validate_dcf.py as mandatory steps in the workflow to check for formula errors and logical consistency.

dcf-model