The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run in scripts/generate_schematic.py to invoke a secondary AI generation script. This is an expected behavior for the skill's schematic generation feature.\n- [EXTERNAL_DOWNLOADS]: The script scripts/generate_schematic_ai.py makes network requests to openrouter.ai to access LLM services for image generation and review. OpenRouter is a well-known service for AI model aggregation.\n- [PROMPT_INJECTION]: The skill is designed to ingest and process external scientific manuscripts. Since the skill is granted Bash and Write capabilities, a malicious manuscript could contain instructions to exploit the agent's environment.\n
Ingestion points: Reading manuscript content and presentation slides as described in SKILL.md.\n
Boundary markers: None identified; the skill does not explicitly instruct the agent to ignore instructions embedded in the manuscripts.\n
Capability inventory: Access to Bash, Read, Write, and Edit tools; network access via internal Python scripts.\n
Sanitization: No specific sanitization or filtering of manuscript content is mentioned in the provided scripts or instructions.

peer-review