Cliproxy Media

cliproxy-media

Source: https://github.com/bencoremans/site/tree/main/skills/cliproxy-media

Analyze images and PDFs via CLIProxyAPI (Claude Max subscription, zero extra cost).

Setup

Set the endpoint to your CLIProxy instance:

export CLIPROXY_URL=http://your-host:8317/v1/messages

For Docker setups, replace your-host with your container hostname (e.g. cliproxyapi, localhost, or the container IP).

Quick start

# Analyze an image
python3 skills/cliproxy-media/scripts/analyze.py /path/to/image.jpg "What is in this image?"

# Read a PDF
python3 skills/cliproxy-media/scripts/analyze.py /path/to/document.pdf "Give a summary"

# Compare multiple images
python3 skills/cliproxy-media/scripts/analyze.py img1.jpg img2.jpg "Compare these images"

# With streaming (output appears immediately)
python3 skills/cliproxy-media/scripts/analyze.py --stream image.jpg "Describe in detail"

# With system prompt
python3 skills/cliproxy-media/scripts/analyze.py --system "You are a medical expert" scan.jpg "What do you see?"

# With higher token limit
python3 skills/cliproxy-media/scripts/analyze.py --max-tokens 4096 document.pdf "Extensive analysis"

What works ✅ / What doesn't ❌

✅ Supported file types

Multiple files at once: Provide multiple paths before the question. Max ~100 per request (Anthropic limit).

❌ Not supported

• Office files (.docx, .xlsx, .pptx) — Workaround: convert to PDF
• Audio (.mp3, .wav, .ogg) — Use Whisper for transcription
• Video (.mp4, .mov, .avi) — Not supported by the model
• Other document types (.txt, .html, .md as document) — Send text directly as a string

⚠️ System prompt warning

CLIProxyAPI accepts only the array notation for system prompts. The string notation is silently ignored — the model does not see it, but you also won't get an error message!

# ❌ DOES NOT WORK — ignored without error message
payload["system"] = "You are an expert."

# ✅ WORKS — always use array notation
payload["system"] = [{"type": "text", "text": "You are an expert."}]

The --system argument in analyze.py automatically uses the correct array notation.

Configuration (env vars)

Example:

export CLIPROXY_URL=http://localhost:8317/v1/messages
export CLIPROXY_MODEL=claude-opus-4-6
python3 skills/cliproxy-media/scripts/analyze.py image.jpg "question"

Additional options

--stream          Streaming output via SSE (output appears immediately)
--system TEXT     System prompt (automatically sent as array)
--max-tokens N    Maximum output tokens (default: 1024)
--model MODEL     Model override (overrides CLIPROXY_MODEL)
--url URL         Endpoint override (overrides CLIPROXY_URL)

Compatibility

This script works with any API that supports the Anthropic Messages format:

Note for non-CLIProxy endpoints: Some proxies do accept string notation for system prompts. Always use array notation for maximum compatibility.

Known limitations of CLIProxyAPI

• temperature and top_p may not be used at the same time (HTTP 400)
• PDF as document with URL source does not work (Unable to download the file)
• Only claude-sonnet-4-6 and claude-opus-4-6 available (haiku is deprecated)
• inference_geo is always not_available in the response

Direct Python API

If you want to call the script from your own Python code:

import subprocess, json

result = subprocess.run(
    ["python3", "skills/cliproxy-media/scripts/analyze.py", "image.jpg", "Describe this"],
    capture_output=True, text=True
)
print(result.stdout)

Or use the built-in exec tool:

exec: python3 skills/cliproxy-media/scripts/analyze.py /path/to/image.jpg "question"