The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and inventory external archives (ZIP files) or directory structures using scripts/ingest.py and scripts/reference.py.
Ingestion points: The skill reads from user-supplied archive paths and inspects internal directory structures like keyframes/ to produce inventories and contact sheets.
Boundary markers: There are no explicit instructions for the agent to treat the content of the ingested archives as untrusted or to ignore embedded instructions within processed files.
Capability inventory: The skill utilizes Python scripts (scripts/ingest.py, scripts/reference.py) to perform filesystem operations, including reading source material and writing derived indexes.
Sanitization: No explicit sanitization or validation logic is described for the file names or metadata within the archives, which could be exploited to manipulate agent output or decision-making during the inspection phase.

scan-ingest