SuperAPI AI Directory

photocraft-batch-image-delivery

Design & creation

View repository

Updated 11-01 02:01

About this skill

Batch image delivery

Task starting point

First read the assets and confirm the delivery purpose, target dimensions or duration, scope of modifications, and required formats. Details that can be determined from the task and assets should directly adopt reasonable defaults; only ask about missing conditions that would change the result.

When a visible editor is needed, hand off to the photocraft-editor skill. Standalone installation:npx skills add full-aigc-skills/photocraft-skills --skill photocraft-editor. The currently added skill is a local candidate; before release, use the local companion plugin, and do not treat this installation command as proof of release.

Execution steps

  1. Inventory each image's orientation, dimensions, color space, and alpha channel, and group them by landscape and portrait.
  2. Define output dimensions, cropping or padding rules, naming, and format; first do a trial run with the most difficult sample.
  3. Apply the same rules to each image one by one, handle anomalous dimensions separately, and keep the original files.
  4. Check the exported count against the input count, and generate a file manifest and list of failures.

Contact sheet and thumbnail index page

When landscape and portrait images need to be arranged in selection order with filenames labeled, use thecontact sheet operation manualand thescripts/contact_sheet.pywithin this skill. The manual covers scaling, grid, text baseline, pagination, original image protection, and CLI/MCP mapping; complete a representative sample page check before batch output.

Tool implementation

Use doc_open/doc_new to open or create, doc_inspect to view layers; command_list to query actual commands, command_run to execute; doc_save to save .pcraft, doc_export to export.

First read the actual tool schema, then fill in parameters; use return values for object or document IDs. Queried commands must be available on the current backend. Choose a new output path before saving, and keep the input. If a required algorithm, format, or command does not exist, state the gap; do not pass off a similar name as a substitute.

Result check

  • Do not mistake scaling for enhancing real detail.
  • Check for color cast, transparent backgrounds turning black, incorrect cropping of landscape/portrait images, and overwriting due to duplicate names.
  • Output files can be reopened, and pixel dimensions and encoding meet requirements.

Check actual content against the user's goal, not just process exit codes. For local edits, inspect local details; for layout tasks, inspect the whole page; for time-based media, inspect the start, end, and key segments. For anomalous results, first check the current project to avoid repeating irreversible operations.

Delivery

Return the editable project, actual derived files, and a brief explanation: what modifications were made, which steps were not executed due to current capabilities, and what requires user judgment. List only files that actually exist and have been checked. If the user subsequently requests rework, read this session's source project and modify locally, preserving accepted content.