OCR and Documents
DocumentsOCR and Documents is an Agent Skill for documents published in the agentic-in/elephant-agent repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Recover text from scanned or image-heavy documents before attempting structured analysis or downstream writing tasks.
QUICK START
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/agentic-in/elephant-agent/blob/HEAD/packages/skills/builtin_packages/productivity/ocr-and-documents/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/ocr-and-documents-777b3d91, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
OCR and Documents
Use this skill when a document is not directly machine-readable.
Preferred Flow
- Determine whether the file is text-native or scanned.
- Run OCR first when the text layer is missing or broken.
- Preserve page order, tables, and obvious headings where possible.
- Only after extraction should you summarize or transform the content.
Guardrails
- State clearly when OCR confidence is low.
- Do not fabricate unreadable text.