lightcraft-named-version-restore

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references documentation and license files from the darktable-org and storytold repositories on GitHub to provide context for image processing workflows.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use a dedicated CLI tool (lightcraft-cli) for photo catalog management, image selection, and file export operations.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data retrieved from tool outputs, such as edit history and version metadata.
  • Ingestion points: Untrusted data enters the context through catalog.query, history.list, and develop.get command outputs as specified in references/craft-steps.json.
  • Boundary markers: Operations are constrained by a structured JSON playbook that defines the expected command sequence.
  • Capability inventory: The skill possesses the ability to write files to the file system via app.export and modify photo settings via develop.set across multiple scripts.
  • Sanitization: The skill relies on specific identifiers like photo IDs and version indices to bind operations, providing a layer of validation for external data.
  • [DYNAMIC_EXECUTION]: The skill generates JSON-formatted command strings that are executed by piping them into the LightCraft CLI. These scripts are constructed from known templates in references/craft-steps.json to automate version restoration tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 06:49 PM