How do I install the Gemini3pro skill?

Run npx skillmds@latest add delneg/gemini3pro in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Gemini3pro safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Gemini3pro?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Gemini3pro free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Gemini3pro?

delneg (@delneg) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Gemini3pro by delneg — Agent Skill

Name: Gemini3pro
Author: delneg

Gemini3pro

Query Google Gemini 3 Pro via grsai.com API for text generation and image analysis. Use for text generation, Q&A, summarization, code generation, creative writing, image analysis/vision, complex reasoning, and structured document generation. Triggers on "ask gemini", "use gemini", "query gemini", "analyze this image with gemini", or when a second opinion from another LLM is needed. Optionally accepts an image input for vision tasks.

delneg 331cde0 2 files · 8.7 KB Updated Sep 17, 2026

File contents

Gemini 3 Pro Text & Vision

Usage

Run from the user's current working directory (do NOT cd into the skill directory):

Text generation:

uv run scripts/query_gemini.py --prompt "your question" [--system-prompt "instructions"] [--output "response.md"] [--api-key KEY]

Image analysis (vision):

uv run scripts/query_gemini.py --prompt "describe this image" --image "path/to/image.png" [--output "analysis.md"] [--api-key KEY]

Parameters

--prompt / -p (required): Question, task, or instructions
--image / -i: Path to image for vision analysis (jpg, png, webp, gif)
--system-prompt / -s: System instructions to set context/persona
--output / -o: Save response to file (always prints to stdout too)
--api-key / -k: API key (overrides GRSAI_API_KEY env var)

API Key

Provide via --api-key argument or set GRSAI_API_KEY environment variable.

Troubleshooting

Symptom	Resolution
`Error: No API key provided.`	Set `GRSAI_API_KEY` env var or pass `--api-key`
`Error: Image not found:`	Wrong path or unreadable file; verify the path exists
HTTP 401	Wrong or expired API key
`Unexpected response format`	API error; check prompt for policy violations
`uv: command not found`	Install: `curl -LsSf https://astral.sh/uv/install.sh \| sh`, then restart terminal

For transient errors (HTTP 429, network timeouts), the script retries automatically up to 3 times. If retries are exhausted, surface the error to the user.

Output

Print the response to stdout. Optionally save to file with --output. Do not pipe output unless the user asks.

Examples

uv run scripts/query_gemini.py --prompt "Explain the difference between TCP and UDP in simple terms"

uv run scripts/query_gemini.py --prompt "What objects are in this image?" --image "photo.jpg" --output "analysis.md"

---name: gemini3prodescription: Query Google Gemini 3 Pro via grsai.com API for text generation and image analysis. Use for text generation, Q&A, summarization, code generation, creative writing, image analysis/vision, complex reasoning, and structured document generation. Triggers on "ask gemini", "use gemini", "query gemini", "analyze this image with gemini", or when a second opinion from another LLM is needed. Optionally accepts an image input for vision tasks.---# Gemini 3 Pro Text & Vision## UsageRun from the user's current working directory (do NOT cd into the skill directory):**Text generation:**```bashuv run scripts/query_gemini.py --prompt "your question" [--system-prompt "instructions"] [--output "response.md"] [--api-key KEY]```**Image analysis (vision):**```bashuv run scripts/query_gemini.py --prompt "describe this image" --image "path/to/image.png" [--output "analysis.md"] [--api-key KEY]```## Parameters- `--prompt` / `-p` (required): Question, task, or instructions- `--image` / `-i`: Path to image for vision analysis (jpg, png, webp, gif)- `--system-prompt` / `-s`: System instructions to set context/persona- `--output` / `-o`: Save response to file (always prints to stdout too)- `--api-key` / `-k`: API key (overrides `GRSAI_API_KEY` env var)## API KeyProvide via `--api-key` argument or set `GRSAI_API_KEY` environment variable.## Troubleshooting| Symptom | Resolution ||---------|------------|| `Error: No API key provided.` | Set `GRSAI_API_KEY` env var or pass `--api-key` || `Error: Image not found:` | Wrong path or unreadable file; verify the path exists || HTTP 401 | Wrong or expired API key || `Unexpected response format` | API error; check prompt for policy violations || `uv: command not found` | Install: `curl -LsSf https://astral.sh/uv/install.sh \| sh`, then restart terminal |For transient errors (HTTP 429, network timeouts), the script retries automatically up to 3 times. If retries are exhausted, surface the error to the user.## OutputPrint the response to stdout. Optionally save to file with `--output`. Do not pipe output unless the user asks.## Examples```bashuv run scripts/query_gemini.py --prompt "Explain the difference between TCP and UDP in simple terms"``````bashuv run scripts/query_gemini.py --prompt "What objects are in this image?" --image "photo.jpg" --output "analysis.md"```

Frequently asked questions