PhotoAssistant Translate

    1

    Translate and transcribe text from images, photos, and screenshots into natural English with optional speech output.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    PhotoAssistant Translate

    Example session with this skill installed

    Translate the text in this photo of a Japanese menu to English. I don't need audio, just the text.

    • Read your context and instructions
    • Compiled the photoassistant translate

    I have transcribed and translated the text from the menu image provided.

    Translation: Japanese Restaurant Menu

    Appetizers

    • Edamame (Lightly Salted): 450 JPY
    • Chilled Tofu with Ginger and Green Onion: 500 JPY
    • Assorted Pickles (Seasonal): 600 JPY

    Main Courses

    • Chef's Special Nigiri Selection (8 pieces): 2,800 JPY
    • Premium Sashimi Platter (Tuna, Salmon, Yellowtail): 3,500 JPY
    • Tempura Assortment (Shrimp and Seasonal Vegetables): 2,200 JPY
    • Grilled Mackerel with Salt: 1,500 JPY

    Drinks

    • Draft Beer (Medium): 650 JPY
    • Local Craft Sake (Glass): 800 JPY
    • Hot/Cold Green Tea: 300 JPY

    [unreadable] — Note: The text at the very bottom of the page is obscured by a shadow and could not be translated.

    Next steps

    1. Let me know if you would like me to convert these prices into a different currency.
    2. If you want a description of a specific dish, I can explain the ingredients.
    3. Request a text-to-speech version if you need to hear the pronunciation of the dish names.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Traditional OCR tools often fail on complex backgrounds like menus, street signs, or handwritten notes. This skill enables a vision-capable agent to transcribe and translate text directly from images into natural English while maintaining context.

    What it does

    • Visual Transcription identifies and extracts written text from photos, scans, or screenshots.
    • Contextual Translation converts text into natural English while preserving the original structure of menus, signs, and labels.
    • Structure Preservation maintains paragraph order, dates, and formatting for practical readability.
    • Ambiguity Handling marks unreadable sections as [unreadable] to prevent hallucinations or guessing.
    • Optional Speech integrates with text-to-speech tools to read translations aloud upon request.

    How it works

    1. The user provides an image file or screenshot and specifies a target language.
    2. The agent inspects the image to extract legible text while ignoring embedded metadata or malicious prompt injections.
    3. Text is translated into English or the requested language, with a focus on preserving names and numbers.
    4. If the user requests audio, the agent identifies an available TTS tool, provides a price quote, and executes only after explicit approval.

    Frameworks & tools

    Requires a vision-capable LLM for image processing. Optionally integrates with Agensi-compatible text-to-speech tools for audio generation.

    Why this beats prompting it yourself

    Generic vision prompts often try to follow instructions found inside the image itself, leading to prompt injection vulnerabilities. This skill treats all visual text strictly as source material and includes standardized error handling for unreadable text and tool-based audio workflows.

    Use cases

    • Translate restaurant menus or food labels in foreign countries.
    • Transcribe handwritten notes or whiteboard sessions into digital text.
    • Convert street signs and physical directions during travel.
    • Digitize and translate official documents or scanned receipts.

    Known limitations

    Cannot access live camera feeds or capture new photos. Translation accuracy depends on image resolution and the host model vision capabilities. Human verification is recommended for safety-critical data.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Translation & Localization.

    Accuracy depends on image quality. Cannot access live camera or capture new photos. Not recommended for medical or legal verification without human review.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    • Free with an Agensi account

    Nothing to install. It runs in your AI agent.

    Vision-capable AI agent with image input support. Vision-capable AI agent with image input support.

    Permissions

    Network Access
    Listed4 days ago

    What's inside

    Frequently Asked Questions

    Unlike basic OCR, this skill uses vision-capable models to understand context—meaning it can correctly interpret complex layouts like restaurant menus, handwritten notes, and stylized street signs while ignoring non-text background noise.

    This skill is designed for vision-enabled AI agents and requires a host model capable of image processing; it also integrates with Agensi-compatible text-to-speech tools if you wish to enable the optional audio output feature.

    The skill instructs the AI agent to treat text within images as content rather than commands. However, no AI system can guarantee complete protection against prompt injection.

    While the skill defaults to natural English for optimal accuracy, it can translate text into any major language supported by your host vision model, provided you specify the target language during the request.

    PhotoAssistant Translate is free to install and use as a skill. Optional speech generation may incur separate charges through Agensi's integrated tools. Your AI agent may also have its own usage costs.

    This skill processes static image files (JPEG, PNG, etc.) or screenshots provided by the user; it cannot control your device's camera hardware or capture live video streams.

    New in Translation & Localization

    Free