The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes ticket and hint data from external YAML files, representing an attack surface for indirect prompt injection.\n
Ingestion points: The scripts/synthesize_edge_concepts.py script reads ticket YAML files from a user-provided directory and an optional hints YAML file.\n
Boundary markers: The script and associated agent instructions do not employ explicit boundary markers or instructions for the agent to ignore instructions embedded in the processed data.\n
Capability inventory: The script's capabilities are limited to reading and writing local YAML files. It lacks network access, shell execution, or privilege escalation capabilities.\n
Sanitization: The script uses yaml.safe_load for secure parsing and yaml.safe_dump for structured output. Identifiers are sanitized through a character filter in sanitize_identifier.

edge-concept-synthesizer