The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands via the openbkn CLI tool, specifically for tasks such as openbkn bkn push, openbkn bkn pull, and openbkn bkn validate.- [EXTERNAL_DOWNLOADS]: The skill references the @openbkn/bkn-sdk package from the NPM registry. This is a vendor-owned resource (openbkn-ai) necessary for the skill's operational workflow.- [INDIRECT_PROMPT_INJECTION]: The skill architecture creates a surface for indirect prompt injection by processing and generating BKN definition files (Markdown and YAML) from potentially untrusted inputs.
Ingestion points: Data is ingested through local file reading and via the openbkn bkn pull command which retrieves existing definitions.
Boundary markers: The templates do not currently define explicit boundary markers or instruction-ignoring directives for generated content.
Capability inventory: The skill facilitates file system modifications and the execution of external CLI tools.
Sanitization: There are no explicit instructions provided to sanitize or validate user-supplied requirements before they are incorporated into the generated BKN files.

create-bkn