The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it retrieves data from the internet while maintaining shell execution and file modification tools.
Ingestion points: External data is ingested via the mcp__firecrawl__firecrawl_search and WebFetch tools.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the skill's logic for processing retrieved web data.
Capability inventory: The skill has access to Bash (permitting python and pip execution), Write, and Edit operations across its reference implementations.
Sanitization: There is no evidence of sanitization, validation, or filtering of content retrieved from external sources before it is used to inform agent decisions or code execution.
[COMMAND_EXECUTION]: The skill is configured with Bash access to allow the agent to execute Python code and manage environment dependencies. It references the installation of several specialized robotics and computer vision libraries, such as pupil-apriltags and ultralytics, which are required for its stated functionality.

drone-cv-expert