Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/curiosity-ai-bv/localocr/gemini-mdgit clone --depth 1 https://github.com/Curiosity-Ai-BV/localOCRWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/curiosity-ai-bv/localocr/gemini-md)<a href="https://agentmods.dev/instructions/curiosity-ai-bv/localocr/gemini-md"><img src="https://agentmods.dev/badge/instructions/curiosity-ai-bv/localocr/gemini-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00759 | $0.00759 |
| Opus 5 | $0.00380 | $0.00380 |
| Sonnet 5 | $0.00152 | $0.00152 |
| Haiku 4.5 | $0.00076 | $0.00076 |
Grade A, and why
localOCR GEMINI.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Curiosity AI Scans (localOCR)
This project is a local, private AI vision tool that leverages Ollama to analyze images and PDFs. It provides both a web-based UI (Streamlit) and a command-line interface for batch processing.
Project Overview
- Goal: To provide a secure, local way to extract information from visual documents and images without sending data to the cloud.
- Core Technology: Uses local vision models (like Gemma 3 12B, Llama 3.2 Vision) running via Ollama.
- Key Features:
- General Description: Generates textual descriptions of images.
- Structured Extraction: Extracts specific fields (e.g., Invoice Number, Date, Total) into JSON/CSV formats.
- PDF Support: Renders and processes PDF pages individually or as a document.
- Dual Interface: Interactive Streamlit UI and headless CLI.
Architecture
The codebase is modularized into the following components:
app.py: The main entry point for the Streamlit web interface. Handles UI layout, state management, and user interaction.cli.py: The command-line interface for headless batch processing. Supports concurrency and rate limiting.core/: Contains the business logic.pipeline.py: Orchestrates the image/PDF processing flow (resize -> encode -> infer).image_utils.py: Handles image resizing and base64 encoding.pdf_utils.py: Manages PDF loading and page rendering using PyMuPDF.json_extract.py: Logic for parsing structured data from model outputs.templates.py: Manages prompt templates.
adapters/: Interfaces with external services.ollama_adapter.py: Handles communication with the local Ollama instance (model listing, chat API).
ui/: Helper modules for the Streamlit UI (e.g., export functionality).utils/: Shared types and small utility functions.
Setup and Usage
Prerequisites
- Python: 3.9 - 3.12
- Ollama: Must be installed and running (
ollama serve). - Models: Vision models must be pulled (e.g.,
ollama pull gemma3:12b).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 74 lines · 759 tokens per session scan A 79552667cf8b
localOCR GEMINI.md is an instructions file published in the GitHub repository Curiosity-Ai-BV/localOCR (148 stars, last pushed 3mo ago), licensed MIT. It adds 759 tokens to every session, about $0.0038 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other instructions, from other repositories
PixelRAG CLAUDE.md
Instructions for StarTrail-org/PixelRAG, covering pixelrag, layout and conventions.
pdf-inline-image-rag-mcp AGENTS.md
AGENTS.md instructions for Joncallim/pdf-inline-image-rag-mcp, covering pdf inline image rag mcp agent guide, invariants, repository map, focused workflow and validation.
pdf-rag CLAUDE.md
Claude Code instructions for abdellahiheiballa/pdf-rag, covering foundational rules, our relationship, proactiveness, designing software and test driven development (tdd).
invoice-builder copilot-instructions.md
Copilot instructions for piratuks/invoice-builder, a project described as: Invoice and quotation builder desktop app with PDF export, designed for small businesses and freelancers. Create, manage, and export invoices and quotes easily using a local database in an Electron-based app.
docs-to-pdf CLAUDE.md
Claude Code instructions for jean-humann/docs-to-pdf, covering claude development guide for docs-to-pdf, project overview, development environment setup, using mise (recommended) and install mise (if not already installed).
PDF-Writer CLAUDE.md
Instructions for galkahana/PDF-Writer, covering claude code context - pdf-writer development guide, project overview, coding standards discovered, project structure and key components.