The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the Bash tool to run local scripts such as scripts/generate_schematic.py and skills/scientific-slides/scripts/pdf_to_images.py. Specifically, the schematic generation takes a user-provided diagram description directly as a command-line argument, which is a potential vector for shell command injection.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by design.
Ingestion points: The skill is intended to ingest and analyze untrusted external content, including scientific manuscripts, grant proposals, and PDF presentations.
Boundary markers: The instructions do not define boundary markers or delimiters to separate the manuscript content from the agent's instructions.
Capability inventory: The skill has access to powerful tools including Bash (for executing scripts), Read, Write, and Edit, which could be exploited if an ingested document contains malicious instructions.
Sanitization: There is no evidence of input validation or sanitization before the external content is processed or used in script arguments.

peer-review