AI Multimodal
Gemini-powered audio/image/video/PDF analysis + image gen via batch scripts
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 20, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · aa33d45
Fetched the repo tree and the raw SKILL.md: it's one skill among ~40 in this monorepo, self-contained under .claude/skills/ai-multimodal/. Verified all three referenced scripts (gemini_batch_process.py, media_optimizer.py, document_converter.py) plus tests and requirements.txt exist; four reference docs exist but references/document-extraction.md (linked in Reference Navigation) 404s. Security clean: scripts only read GEMINI_API_KEY from env/.env and call the official google-genai SDK, no curl|sh/base64/exfiltration. TRIGGER (5/5): would load — "Transcribe this MP3 with timestamps", "Extract the tables from this PDF as JSON", "Generate an image of a futuristic city at sunset, 16:9"; would NOT load — "Write a blog post about multimodal AI trends", "Set up my GitHub Actions CI pipeline". OUTPUT: NOT measured (not a measured loss) — runtime needs the google-genai SDK + a live Gemini API key I can't provision, so no baseline-vs-skill comparison was possible; scored 5 as unmeasured-neutral. INSTALL docked for external SDK+API-key dependency (inert on copy alone) and the one missing reference doc; DOCS docked for that dead link and unverifiable Gemini capacity claims.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 5/5
- Output vs. baseline 5/10
- Docs & honesty 4/5
What AI Multimodal does
Wraps Google Gemini's multimodal API in three Python batch scripts to transcribe audio, analyze images/video, extract structured data from PDFs, and generate images. Triggers when working with media files, screenshots, PDF extraction, or text-to-image tasks. Requires installing the google-genai SDK and a GEMINI_API_KEY.
How to install AI Multimodal
Copy .claude/skills/ai-multimodal/ into your project's .claude/skills/, then `pip install google-genai python-dotenv pillow` and set GEMINI_API_KEY (env or .env). Get a key at https://aistudio.google.com/apikey.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger AI Multimodal
-
/ai-multimodalGemini-powered audio/image/video/PDF analysis + image gen via batch scripts
It also activates on plain-language prompts like these:
-
Transcribe and summarize this audio recording -
Extract tables from this scanned PDF -
Detect objects and answer questions about this image
Frequently asked questions
- Is the AI Multimodal skill free?
- Yes. The skill itself is free from mrgoonie/claudekit-skills. SkillProof publishes the install command and an independent test verdict at no cost.
- Does AI Multimodal work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 20, 2026. Verdict: Works with setup. Fetched the repo tree and the raw SKILL.md: it's one skill among ~40 in this monorepo, self-contained under .claude/skills/ai-multimodal/. Verified all three referenced scripts (gemini_batch_process.py, media_optimizer.py, document_converter.py) plus tests and requirements.txt exist; four reference docs exist but references/document-extraction.md (linked in Reference Navigation) 404s. Security clean: scripts only read GEMINI_API_KEY from env/.env and call the official google-genai SDK, no curl|sh/base64/exfiltration. TRIGGER (5/5): would load — "Transcribe this MP3 with timestamps", "Extract the tables from this PDF as JSON", "Generate an image of a futuristic city at sunset, 16:9"; would NOT load — "Write a blog post about multimodal AI trends", "Set up my GitHub Actions CI pipeline". OUTPUT: NOT measured (not a measured loss) — runtime needs the google-genai SDK + a live Gemini API key I can't provision, so no baseline-vs-skill comparison was possible; scored 5 as unmeasured-neutral. INSTALL docked for external SDK+API-key dependency (inert on copy alone) and the one missing reference doc; DOCS docked for that dead link and unverifiable Gemini capacity claims.
- What is the AI Multimodal SkillProof Score?
- 7.2/10 — installs cleanly 4/5, triggers reliably 5/5, output vs. baseline 5/10, docs & honesty 4/5.
- How do I install AI Multimodal?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use AI Multimodal with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.