The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the research-lookup tool to populate report sections. While it lacks explicit boundary markers or sanitization for this external content, the risk is mitigated by the structured LaTeX template format.
Ingestion points: research-lookup results gathered in Phase 1 and integrated into LaTeX drafts.
Boundary markers: Absent; data is interpolated directly into the .tex files.
Capability inventory: Read, Write, Edit, and Bash tools are used to manage files and compile the LaTeX document into PDF format using xelatex.
Sanitization: No specific sanitization of external market data is mentioned before inclusion in reports.
[COMMAND_EXECUTION]: The skill uses Bash to execute LaTeX compilation commands (xelatex, bibtex) and to run local Python scripts for visual generation. The scripts/generate_market_visuals.py script uses subprocess.run to call internal tools (scientific-schematics, generate-image) with arguments derived from the report topic. These operations are essential for the skill's primary purpose and do not represent unauthorized command execution.

market-research-reports