OpenParable
Skip to content
← Back to skills

Gemini Feedback

Documents
gemini-feedback

Gemini Feedback is an Agent Skill for documents published in the natolambert/rlhf-book repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Get feedback from Gemini API on a diagram image for textbook quality review.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/natolambert/rlhf-book/blob/HEAD/.claude/skills/gemini-feedback/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/gemini-feedback, 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.

Gemini Feedback

Get feedback from Gemini API on a diagram image.

Usage

/gemini-feedback <path-to-image> [context]

Instructions

When this command is invoked:

  1. Read the image file at the provided path
  2. Call the Gemini API to get feedback on the diagram
  3. Use this Python snippet to call Gemini (uses the newer google-genai package):
import base64
import os
from google import genai

client = genai.Client(api_key=os.environ['GEMINI_API_KEY'])

with open('<IMAGE_PATH>', 'rb') as f:
    img_bytes = f.read()

prompt = '''Review this diagram for a textbook. Be concise and specific.

Context: <CONTEXT>

Please provide:
1. Overall assessment (1-2 sentences)
2. Specific issues to fix (be detailed about visual problems like alignment, overlapping, etc.)
3. Suggestions for improvement
'''

response = client.models.generate_content(
    model='gemini-2.5-flash',
    contents=[
        prompt,
        genai.types.Part.from_bytes(data=img_bytes, mime_type='image/png'),
    ],
)
print(response.text)
  1. Replace <IMAGE_PATH> with the actual path provided
  2. Replace <CONTEXT> with any context provided, or use "Technical diagram for ML/AI textbook"
  3. Ensure GEMINI_API_KEY is set in environment
  4. Run using: uv run python -c "..."
  5. Present the feedback to the user

Example

/gemini-feedback diagrams/generated/png/tool_use_generation.png "Diagram showing tool use interleaving in LLM generation"