The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes af commands via the uvx runner to perform Airflow operations such as triggering DAGs and retrieving logs.
[EXTERNAL_DOWNLOADS]: The skill fetches the astro-airflow-mcp package from the vendor's repository at runtime. This is a trusted resource owned by the author 'astronomer'.
[REMOTE_CODE_EXECUTION]: The agent uses uvx to download and run the vendor-provided CLI tool. This is standard behavior for this vendor's management tools.
[CREDENTIALS_UNSAFE]: The skill enables the agent to view Airflow connection details and variables through af config connections and af config variables. This is necessary for identifying configuration-related failures in the DAG environment.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface through the ingestion of external task logs and diagnostics.
Ingestion points: Data enters the context via af tasks logs and af runs diagnose commands.
Boundary markers: No specific delimiters are defined to separate log content from agent instructions.
Capability inventory: The agent maintains shell command capabilities via uvx throughout the workflow.
Sanitization: External log data is not sanitized before being processed by the agent.

testing-dags