The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the use of the frameproof-video-qa package via npx, which downloads and executes the tool from the npm registry. This is consistent with the skill's purpose as a documentation layer for the utility and utilizes a well-known package registry.
[COMMAND_EXECUTION]: The skill provides instructions for executing various CLI commands to perform video validation, metadata scanning, and automated repair workflows (e.g., npx frameproof-video-qa verify, npx frameproof-video-qa scan).
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data including video files, subtitle files, and JSON manifests to detect defects.
Ingestion points: SKILL.md and references/verify-and-repair.md describe commands that read video files (--video), subtitle files (--captions), and manifest files (--manifest).
Boundary markers: No specific boundary markers or instructions to ignore embedded content are implemented for the ingested data.
Capability inventory: The skill utilizes npx to execute a scanner that performs file system reads and writes artifacts (JSON, HTML, SARIF).
Sanitization: The tool performs structured analysis of file metadata and contents, though no explicit sanitization of the input data is mentioned in the documentation.

video-qa-frameproof