The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external shot briefs, manifests, and quality control files which could potentially contain malicious instructions.
Ingestion points: Workflow reads TAKE_SELECTION.md, MEDIA_QC.md, take_qc.jsonl, and adjacent shot records.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions within the ingested content.
Capability inventory: The skill has the ability to write to take_manifest.jsonl, selections.json, and execute the project-impact tool.
Sanitization: The workflow lacks explicit validation or sanitization steps for the data processed from files.

take-selection