The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external prompt data stored in JSON files within the extracted_results/ directory to generate Markdown documentation. This ingestion of untrusted content represents a surface for indirect prompt injection, where instructions embedded in the analyzed data could influence the agent's behavior during the generation of 'knowledge cards'.
Ingestion points: The load_prompts function in xray_helper.py reads data from files matching *_extracted.json.
Boundary markers: The skill's output templates in skill.md do not utilize explicit delimiters or warnings to ignore instructions within the ingested data.
Capability inventory: The skill is capable of reading from and writing to the local filesystem using xray_helper.py.
Sanitization: The skill performs no validation, escaping, or filtering of the ingested JSON content before interpolating it into Markdown templates.

prompt-xray