Before executing the specific method, read Craft scenario commands and parameters, and execute according to the object selection, session, input replacement, and result checking therein.
Task Starting Point
First read the materials, confirm the delivery purpose, target size or duration, modification scope, and required format. Directly adopt reasonable defaults for details that can be determined from the task and materials; only ask about missing conditions that would change the result.
When a visible editor is needed, hand off to the pdfcraft-editor skill. Standalone installation:npx skills add full-aigc-skills/pdfcraft-skills --skill pdfcraft-editor. The currently added skill is a local candidate; use the local companion plugin before release, and do not treat this installation command as proof of publication.
Execution Steps
- List the specific fields and page numbers, and save an independent copy.
- Use genuine removal or a verified rasterization process; do not merely draw black rectangles.
- Check whether the text layer, annotations, attachments, and metadata still contain content.
- Reopen the output and attempt to search, copy, and extract the target text.
Verify content deletion, fake-redaction audit, and hidden-information checks separately; when metadata or attachments need to be handled, operate within the scope authorized by the user.
Tool Implementation
The upstream tool name belongs to PdfCraft. doc_open/doc_create return a doc handle; subsequent doc_info, doc_save, and page_render use the same doc. PDF page numbers start at 1, and parameters are subject to the actual schema.
First read the actual tool schema, then fill in the parameters; use returned values for object or document IDs. The queried commands must be available on the current backend. Choose a new output path before saving, and preserve the input. If the required algorithm, format, or command does not exist, explain the gap; do not pass off a similar name as a substitute.
Result Checking
- If the current capability cannot verify thorough removal, stop and explain.
- Check whether the redacted area is complete and does not cover unrelated content.
- Deliver the cleaned new file without overwriting the original.
Check the actual content against the user's goal, not just the process exit code. For local edits, inspect local details; for layout tasks, inspect the full page; for time-based media, inspect the start, end, and key segments. For abnormal 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 capability, and what requires user judgment. List only files that actually exist and have been checked. If the user subsequently requests rework, read this source project and make local modifications, preserving the accepted content.
Skill Store Scenario Experience
When encountering the deletion of sensitive content rather than covering it with black boxes, readStore Workflow Adaptation, and use the corresponding Craft method and actual result checking. Upstream skills are not runtime dependencies.