The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Company filings (BamSEC, SEC EDGAR), investor presentations, corporate websites, and news/press releases (SKILL.md).
Boundary markers: None identified; external content is not wrapped in protective delimiters.
Capability inventory: The skill executes shell commands (soffice, pdftoppm) and performs web research (SKILL.md).
Sanitization: No validation or sanitization of external data is specified before it is processed into slides.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to facilitate document conversion and visual verification.
Evidence: The workflow requires running soffice --headless --convert-to pdf and pdftoppm to generate slide previews.
[DATA_EXFILTRATION]: While primary data flow is for research, the skill accesses external financial platforms. This is documented as a standard function of the skill's purpose for financial analysis.

fsi-strip-profile