The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and analyze external React component files (.tsx), which introduces a surface for indirect prompt injection. Malicious instructions could be embedded within the source code comments or string literals of the input file to influence the agent's behavior during the analysis or subsequent implementation phases.
Ingestion points: The contents of the React component file specified by the component-path parameter are parsed and analyzed.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between the input component data and its own task instructions.
Capability inventory: The skill requires the ability to read local files and write results to the filesystem (creating .temp.json and .temp.md files).
Sanitization: The skill does not define any sanitization or filtering logic for the data extracted from the source files before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill documentation references a prerequisite tool named skeleton to perform AST (Abstract Syntax Tree) analysis. The skill does not provide the source, installation scripts, or integrity checks for this tool, making it an unverifiable external dependency.
[COMMAND_EXECUTION]: The operational workflow requires the agent to execute the skeleton tool using a user-provided file path. This pattern poses a risk of command injection if the agent fails to properly sanitize the component-path input before passing it to a shell environment.

generate-preview