The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions involve executing a shell command via uv run researchkit. The user-supplied question is interpolated directly into the command line, which presents a risk of command injection if shell metacharacters are included in the input.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests untrusted external data which could influence agent behavior.
Ingestion points: The QUESTION argument and the --context-file notes.md file (SKILL.md).
Boundary markers: No delimiters or "ignore embedded instructions" markers are defined for the ingested data.
Capability inventory: The skill utilizes subprocess execution to run the researchkit tool (SKILL.md).
Sanitization: There are no explicit instructions for sanitizing, escaping, or validating the input provided in the QUESTION or the context file.

advise-max