The Agent Skills Directory

[SAFE]: The skill instructions are focused entirely on financial analysis and report generation. There are no signs of command injection, credential harvesting, or unauthorized network activity.
[INDIRECT_PROMPT_INJECTION]: The skill workflow involves processing external data. Evidence Chain: 1. Ingestion points: Financial filings (10-K, 10-Q), transcripts, and shareholder letters referenced in banks.md, energy.md, and software-and-internet.md. 2. Boundary markers: Not explicitly specified in the skill instructions. 3. Capability inventory: Analytical reasoning and report generation using internal docx, pptx, and xlsx formatting skills. 4. Sanitization: The instructions mitigate risk by emphasizing the classification of figures as 'sized' or 'extrapolated' and requiring 'company claim' labels for non-accounting standards data.

sector-overview