The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform specific shell operations to generate schematics and compile LaTeX reports. Evidence includes the usage of python scripts/generate_schematic.py, python scripts/generate_image.py, and xelatex report.tex within the drafting and formatting instructions.
[INDIRECT_PROMPT_INJECTION]: The skill relies on the research-lookup tool to ingest external literature, creating a vulnerability surface where instructions hidden in research papers could influence the agent's behavior. 1. Ingestion points: Data retrieved via the research-lookup tool (SKILL.md). 2. Boundary markers: None specified in the instructions to separate fetched data from the agent's core logic. 3. Capability inventory: Subprocess execution via Bash (python, xelatex), and file system modification via Write and Edit. 4. Sanitization: No specific sanitization or validation of the fetched external research data is described.

scientific-writing