← Back to Skills Marketplace
Ollama OCR
by
hongjiahao371-pixel
· GitHub ↗
· v1.0.0
· MIT-0
283
Downloads
0
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install ollama-ocr
Description
Use Ollama's vision/OCR models to recognize text from images. Supports glm-ocr, llava, moondream, and llama3.2-vision models. Ideal when you need local offli...
Usage Guidance
This skill appears to do exactly what it claims: read an image and send it to an Ollama instance for OCR. Before installing/using, verify you run a trusted Ollama instance at the configured host/port (default 172.17.0.2:11434) — if that endpoint is remote or controlled by someone else, your images would be sent off-host. Note the script hardcodes OLLAMA_HOST/PORT; if your Ollama runs elsewhere you must edit the script (or add environment-variable support) to point it to your instance. No credentials are requested by the skill, but be mindful of the images you pass in (they are transmitted to the configured Ollama HTTP endpoint).
Capability Analysis
Type: OpenClaw Skill
Name: ollama-ocr
Version: 1.0.0
The ollama-ocr skill is a straightforward implementation for local image text recognition using Ollama. The Python script (ollama_ocr.py) uses standard libraries to communicate with a local API endpoint (172.17.0.2) and contains no evidence of data exfiltration, malicious execution, or obfuscation.
Capability Assessment
Purpose & Capability
Name/description (local Ollama OCR) match the code and SKILL.md: the script encodes an image and POSTs it to an Ollama /api/chat endpoint for model-based OCR. No unrelated services, binaries, or credentials are requested.
Instruction Scope
SKILL.md and the Python script limit actions to reading the specified image file and sending it to the Ollama HTTP API; there are no instructions to read other files, environment secrets, system configs, or transmit data to unexpected endpoints. The SKILL.md correctly documents the default endpoint used by the script.
Install Mechanism
This is an instruction-only skill with a small Python script and no install spec. Nothing is downloaded or written to disk by an installer; risk from installation is minimal.
Credentials
The skill requests no environment variables, credentials, or config paths. It relies on an accessible Ollama HTTP endpoint; this is proportional to its purpose. (Note: the endpoint is hardcoded in the script rather than read from an environment variable.)
Persistence & Privilege
The skill does not request persistent/autonomous privileges, does not set always:true, and does not modify other skills or system-wide settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ollama-ocr - After installation, invoke the skill by name or use
/ollama-ocr - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release - local OCR with Ollama vision models
Metadata
Frequently Asked Questions
What is Ollama OCR?
Use Ollama's vision/OCR models to recognize text from images. Supports glm-ocr, llava, moondream, and llama3.2-vision models. Ideal when you need local offli... It is an AI Agent Skill for Claude Code / OpenClaw, with 283 downloads so far.
How do I install Ollama OCR?
Run "/install ollama-ocr" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Ollama OCR free?
Yes, Ollama OCR is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Ollama OCR support?
Ollama OCR is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Ollama OCR?
It is built and maintained by hongjiahao371-pixel (@hongjiahao371-pixel); the current version is v1.0.0.
More Skills