The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs legitimate command execution to support its primary function.
Evidence: The script scripts/generate_market_visuals.py uses subprocess.run to call external visual generation tools (scientific-schematics and generate-image).
Evidence: SKILL.md provides instructions for the agent to use xelatex and bibtex to compile the generated reports into PDFs.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by interpolating user-controlled data into prompts used for subsequent tool calls.
Ingestion points: The topic provided by the user (via the --topic argument in the generation script or as a placeholder in manual research commands) enters the agent's context in SKILL.md and scripts/generate_market_visuals.py.
Boundary markers: There are no explicit boundary markers or delimiters used when the topic is inserted into prompt templates for research-lookup or the visual generation tools.
Capability inventory: The skill has access to Bash, Read, Write, and Edit, allowing it to execute scripts, create files, and modify existing project structures based on its research results.
Sanitization: No input validation or escaping is performed on the user-provided topic before it is passed to other tools, creating a risk that malicious input could influence the behavior of the research or image-generation agents.

market-research-reports