Extracts precise visual details from images, transcribes audio/video, converts documents to markdown, and generates images.
- GitHub stars
- 126
- Files
- 9
- Last commit
- May 16, 2026
- License
- MIT
- Allowed tools
- —
What does the media-processor skill do?
media-processor extracts precise visual details from images, transcribes audio/video, converts documents to markdown, and generates images.
Summary generated from the skill's SKILL.md.
About the media-processor skill
Extracting precise visual details from UI mockups or dense screenshots is what this skill enables for developers and designers. It provides specialized tools to identify exact hex colors, spacing in pixels, font sizes, and component hierarchies from images. When an agent needs to audit a user interface against a design or compare screenshots for visual regressions, it can load these tools to produce highly accurate technical specifications.
The skill utilizes three primary scripts located in the scripts directory. The gemini_batch_process.py script handles image analysis, audio and video transcription, and data extraction from PDFs. The document_converter.py script transforms PDF, DOCX, XLSX, and PPTX files into markdown format. For creative tasks, image_gen.py allows for image generation and editing, though this requires a paid plan. All operations rely on a GEMINI_API_KEY and can be fine-tuned using model overrides like gemini-3.1-pro-preview for complex visual tasks.
What an agent can do with it
- 1Extract UI specificationsAsk the agent to identify exact colors and spacing from a design mockup for implementation.
- 2Convert documents to markdownRequest the conversion of complex PDF or Office files into structured markdown text.
- 3Perform visual regression auditsHave the agent compare a live screenshot against an original design to find inconsistencies.
- 4Transcribe multimedia filesUse the tool to turn audio or video content into written text.
Written from the skill's SKILL.md. Every use case is checked against a quote from it.
How to install the media-processor skill
Each command copies the skill folder into the client's skills directory. Restart the client or start a new session to load it.Install the media-processor skill in Claude Code
~/.claude/skills/media-processor/SKILL.md
$ npx degit avibebuilder/claude-prime/.claude/skills/media-processor ~/.claude/skills/media-processorInstall the media-processor skill in Claude Code (project)
.claude/skills/media-processor/SKILL.md
$ npx degit avibebuilder/claude-prime/.claude/skills/media-processor .claude/skills/media-processorInstall the media-processor skill in Codex
~/.agents/skills/media-processor/SKILL.md
$ npx degit avibebuilder/claude-prime/.claude/skills/media-processor ~/.agents/skills/media-processorInstall the media-processor skill in Gemini CLI
~/.gemini/skills/media-processor/SKILL.md
$ npx degit avibebuilder/claude-prime/.claude/skills/media-processor ~/.gemini/skills/media-processorInstall the media-processor skill in Claude.ai
npx degit avibebuilder/claude-prime/.claude/skills/media-processor media-processor && zip -r media-processor.zip media-processor # claude.ai → Customize → Skills → + → Create skill → Upload a skill → media-processor.zip
Automated checks
Ran Nov 16, 2026Full report →Passed our automated checks. That is not a security audit: read the scripts before you let an agent run them. How checks work
- Source repository linked and reachableavibebuilder/claude-prime/.claude/skills/media-processor is public and reachablegithub.com/avibebuilder/claude-prime/tree/HEAD/.claude/skill ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit 2026-05-16github.com/avibebuilder/claude-prime/commits ↗Pass
- No download-and-run or credential theft in scripts3 scripts read; no download-and-run, remote eval, base64 execution or credential uploadPass
- No hidden instructions in SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/avibebuilder/claude-prime/blob/HEAD/.claude/skill ↗Pass
- Pre-approved tools are scopedNo pre-approved tools; the agent asks before using onegithub.com/avibebuilder/claude-prime/blob/HEAD/.claude/skill ↗Pass
- SKILL.md follows the Agent Skills formatValid name and description; says when to use it; 75 linesgithub.com/avibebuilder/claude-prime/blob/HEAD/.claude/skill ↗Pass
SKILL.md
View on GitHubMedia Processor
Specialized tools for extracting precise visual details (exact colors, spacing, hierarchy), processing audio/video, and generating images.
Tools
All scripts live in scripts/ relative to this skill's directory. They auto-select the best model per task and handle retries, large file uploads, and error reporting.
| Script | Purpose |
|---|---|
gemini_batch_process.py |
Analyze images, transcribe audio/video, extract data from PDFs |
image_gen.py |
Generate and edit images (paid plan required) |
document_converter.py |
Convert PDF, DOCX, XLSX, PPTX to Markdown; extract page ranges and images |
Requires GEMINI_API_KEY in environment or .env in this skill's directory. Run any script with --help for setup details and available parameters.
Quick start — image analysis:
python <skill-dir>/scripts/gemini_batch_process.py \
--files <image-path> \
--task analyze \
--prompt "<tailored prompt>" \
--output <output-path>.md
Prompt Quality Matters
The prompt sent to the processing model is the single biggest factor in output quality. Tailor prompts to what the task actually needs — generic prompts produce generic results.
What makes a good analysis prompt:
- Ask for the specific details the task requires (hex colors, spacing in px, component hierarchy) rather than "describe this image"
- Structure the ask as a numbered list — the model mirrors the structure back, making output easy to parse
- Name the desired output format ("as a markdown table", "as JSON", "as a component tree")
- Include implementation context when relevant ("for React with Tailwind") so the model emphasizes useful details
Example prompt patterns:
UI implementation: "Extract component hierarchy, layout type, exact hex colors, typography (sizes/weights), spacing in px, interactive states, icons and decorative elements"
Chart data: "Extract chart type, axes with units, every data point with exact values, legend entries with colors. Output as a markdown table"
Design review: "Compare this screenshot against the design. Flag differences in spacing, colors, alignment, missing elements, and visual inconsistencies. Note exact values for each discrepancy"
Pasted Images
When a user pastes images in chat, they are auto-saved to:
$CLAUDE_DIR/image-cache/<current_session_id>/<image_number>.png
Use ls "$CLAUDE_DIR/image-cache/" to discover the session ID, then list its contents to find available images.
Model Overrides
Scripts auto-select models per task (see model-routing.md). Override with --model <model-id> when the default isn't enough — for example, --model gemini-3.1-pro-preview for complex visual analysis where the pro model catches more detail than flash.
References
| Reference | When to read |
|---|---|
| api-gotchas.md | Before using image generation, video processing, or raw API calls — prevents common failures |
| model-routing.md | When choosing or overriding the default model for a task |
| media-optimization.md | When files are too large to upload — ffmpeg compression recipes |
Gotchas
- Rate limits — scripts retry up to 3 times with backoff. If still rate-limited after retries, stop and ask the user to check their API key quota or provide a new key.
- Model IDs change — Google frequently rotates preview model IDs. If you get a 404, the model was likely superseded — check the models page for current IDs.
- Safety filters — the API may refuse some content. Report clearly to the user rather than retrying.
- Large files auto-upload — files >20MB automatically use the File API (2GB max, 48h retention). No action needed.
Files in this skill
9media-processor skill questions
How do I install the media-processor skill in Claude Code?
Run npx degit avibebuilder/claude-prime/.claude/skills/media-processor ~/.claude/skills/media-processor. Claude Code loads skills from ~/.claude/skills (all projects) or .claude/skills (one project) and uses a skill when its description matches the task.
Which clients can use the media-processor skill?
Any client that reads the Agent Skills format: Claude Code, Codex, Gemini CLI and claude.ai (upload as a ZIP). Scripts in the skill run only where the client allows code execution.
What happens if I try to upload a very large file?
Files larger than 20MB will automatically use the File API, which supports up to 2GB and retains files for 48 hours.
How do I convert a PDF or DOCX to Markdown?
Use the document_converter.py script, which can convert PDF, DOCX, XLSX, and PPTX files into Markdown and extract images or page ranges.