The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to utilize standard Unix utilities, including wc, grep, awk, and sed, to navigate and extract relevant sections from large AsyncAPI specification files. This approach is a best practice for managing large contexts by reading only the necessary fragments of a document.
[DYNAMIC_EXECUTION]: The skill is designed to generate Drift YAML test definitions. These definitions frequently include executable-type: command and executable-type: lua-function blocks, which are intended to reference user-provided shell scripts (e.g., Python) or Lua functions for test triggers and probes. This is a standard feature of the Drift testing framework and is used here for its intended purpose.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it processes external, user-supplied AsyncAPI specifications.
Ingestion points: The agent reads local files (e.g., asyncapi.yaml) and resolves external references (e.g., ./messages.yaml) provided by the user.
Boundary markers: There are no specific instructions to use delimiters or ignore instructions embedded within the data of the AsyncAPI files.
Capability inventory: The skill utilizes shell command execution via the agent's environment to parse and process the files.
Sanitization: The skill relies on structural parsing of the YAML/JSON content and does not explicitly sanitize the text content of the specification fields.

asyncapi-parser