The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources like tickets and logs which could contain malicious instructions.\n- Ingestion points: Technical data is ingested through user prompts and a JSON file read by the scripts/release_notes_generator.py tool.\n- Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when the agent processes the external content.\n- Capability inventory: The skill employs a Python script with file-read capabilities and leverages the agent's text generation for rewriting tasks.\n- Sanitization: The skill does not perform sanitization of input descriptions to filter out potential injection payloads.

release-notes