The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing the knap CLI tool globally using npm install -g knap or running it via npx knap. This introduces a dependency on an external package from the NPM registry.
[COMMAND_EXECUTION]: The skill utilizes shell commands to perform its core functions, including rendering templates (knap render), batch processing (knap batch), and syntax validation (knap validate). It also suggests piping output from other tools like defuddle.
[DYNAMIC_EXECUTION]: The skill employs a template engine that supports logic tags ({% ... %}) and filters (|), allowing for dynamic content generation based on input data at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its data ingestion pipeline.
Ingestion points: The skill processes data from local JSON/CSV files and external web content via the defuddle parse tool.
Boundary markers: No explicit boundary markers or "ignore instructions" warnings are present in the template examples to prevent the agent from following instructions embedded in the data.
Capability inventory: The skill has the capability to write rendered output to the local file system using the -o (output file) and --output-dir flags.
Sanitization: Although a safe_name filter is suggested for filenames to prevent directory traversal or invalid characters, there is no evidence of sanitization for the main template body content when interpolating external data.

knap