The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the akf command-line interface to perform operations such as metadata stamping, reading, and auditing.
[EXTERNAL_DOWNLOADS]: The documentation instructs users to install the akf package via pip and the akf-format package via npm.
[PROMPT_INJECTION]: The skill processes external files to audit and inspect metadata, creating an indirect prompt injection surface.
Ingestion points: The akf read, akf inspect, and akf audit commands ingest data from user-provided files.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded content within the files being processed.
Capability inventory: The skill executes shell commands (akf) based on file processing results.
Sanitization: The instructions do not specify any sanitization or validation of the ingested metadata before processing.

akf-trust-metadata