compdfkit/compdf-ocr
Recognize and extract text from scanned PDFs and images with ComPDF OCR workflows. Use for OCR, searchable PDF generation, text recognition, table recognition, and scanned-document extraction requests.
npx skills add https://github.com/ComPDFKit/compdf-skills --skill compdf-ocr
ComPDF OCR helps agents and teams unlock text from scanned PDFs, screenshots, photographed files, and other image-based documents. It supports OCR on both PDFs and image files, making it easier to extract readable text, recover structured content, and create searchable outputs for analysis, archiving, and workflow automation.
Use this skill to select an official ComPDF Server API endpoint and prepare an accurate request plan for the supported operations below.
| Operation | Official page or index section |
| --- | --- |
| PDF to editable/searchable PDF | pdf-to-editable-pdf-tool-guide |
| PDF text extraction | pdf-to-txt |
| Image OCR to text | image-to-txt |
| OCR language values | ocr-languages |
Restrict this skill to OCR and OCR-derived text or searchable-PDF output. Use the matching conversion skill for other target formats.
references/endpoint-index.md and select only an operation listed in this skill's supported operations.references/official-api-reference.md. Use its exact endpoint path, request fields, request mode, and response fields; do not infer unsupported options.COMPDF_API_KEY_FILE when set. Otherwise read %USERPROFILE%\.compdf\api_key on Windows, or ~/.config/compdf/api_key on macOS and Linux. Pass the value only as the x-api-key header; never put it in code, logs, examples, or output.Use one local, private key file so later ComPDF tasks do not require pasting an API key into chat. The file must contain only the API key on its first non-empty line. Do not create, commit, upload, or display this file.
When the selected file is absent, unreadable, or empty, direct the user to obtain a key at https://www.compdf.com/compdf-portal/signin?utm_source=github&utm_medium=referral&utm_campaign=compdf_skills_repo_en&ref_platform_id=github_compdfkit_skills_en, save it in the selected file, and retry.
Refresh the local official snapshot before release, then inspect the diff for renamed endpoints, changed fields, and changed enum values:
python scripts/sync_official_api_reference.py
Take compdfkit/compdf-ocr from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.