The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection through its ingestion of external results artifacts. The scripts/render_results_report.py script parses user-provided JSON files and interpolates the values directly into a Markdown report template without escaping or sanitization.\n- Ingestion points: The script scripts/render_results_report.py reads data from a user-specified JSON file path (e.g., data/results.json) via the --json argument.\n- Boundary markers: The script does not implement boundary markers or instructions to the agent to ignore embedded commands within the generated report.\n- Capability inventory: The skill is capable of reading and writing to the file system and executing Python scripts, providing a path for injected instructions to perform further actions.\n- Sanitization: Absent. The renderer does not perform validation or sanitization of the strings extracted from the JSON artifact before including them in the final output document.

results-reporting