Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    music score digitizer

    1

    Professional sheet music digitization workflow using machine-drafting, automated code gates, and AI-led engraving audits.

    Free

    Security scanned

    music score digitizer

    Example session with this skill installed

    Digitize the attached two-page Tchaikovsky arrangement scan. Extract the embedded images from the PDF first, run the accidental carry gate, and provide the MusicXML and a verification checklist.

    • Read your context and instructions
    • Compiled the music score digitizer
    • Generated the document

    I have completed the digitization workflow for the Tchaikovsky arrangement. The source audit confirmed an embedded resolution of 219 DPI (18px staff spacing), which passed the quality gate for high-reliability transcription.

    I extracted the raw images, generated a structured JSON draft, and ran the mandatory validation gates. The check_accidental_carry script identified 4 instances where sharps were correctly carried through the measure despite no printed accidental on subsequent notes. Rhythmic integrity was verified at 100% across all 38 measures.

    Digitization Results

    MetricValueStatus
    Source DPI219 (Embedded)✅ Reliable
    Total Measures38✅ Verified
    Validation GatesBeats, Accidentals, Ties, Barlines✅ Passed
    Output Filesscore.musicxml, score.pdf, score.mp3✅ Generated
    Confidence Avg99.2%✅ High

    Human-Verification Checklist (Priority Subset)

    MeasureDraft ReadingPrinted ScoreTheory OpinionSuggested Action
    7dynamic: pglyph: mpBlurred 'm' in sourceConfirm if Mezzo-Piano
    14E6 (top ledger)E6Ledger-stack rule appliedKeep (Rule 4.1)
    22slur arc( ) glyphsPortato style in metal-typeKeep as semi-legato

    Next steps

    1. Review the mp dynamic marking in measure 7 against your physical copy.
    2. Import the score.musicxml into MuseScore to adjust personal layout preferences.
    3. Verify the Metronome Mark (♩=132) used for the reference MP3 matches your performance intent.

    score.pdf

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

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

    About this skill

    NOTE:This skill requires a vision-capable (multimodal) model, and its output is a high-confidence draft — always have a human proofread the final score against the original scan before use. High-fidelity sheet music digitization often fails due to prohibitive token costs or inaccurate Optical Music Recognition (OMR) on low-resolution scans. This skill implements a "machine-draft, AI-audit" workflow that reduces overhead by 80% while maintaining professional engraving standards.

    What it does

    • Source Auditing extracts embedded originals from PDFs and measures staff-line spacing to determine OCR feasibility.
    • Automated Validation executes scripted gates for beat values, accidental carry-over, barline integrity, and tie/slur pairing.
    • High-Precision Transcription converts scanned imagery into MusicXML, print-quality vector PDFs, and reference MP3s.
    • Engraving Rule Enforcement applies publisher-specific typography rules for ledger lines, notehead seated positions, and beam thickness.
    • Conflict Arbitration uses music theory logic to resolve discrepancies between machine drafts and printed sources.

    How it works

    1. Audit phase: Run extract_pdf.py and staff-spacing checks to verify if the source resolution meets the 16px line-spacing threshold.
    2. Drafting phase: Generate a structured JSON draft using Audiveris or internal chord scanners, writing confidence scores to disk.
    3. Validation phase: Execute mandatory Python gates to flag rhythmic errors or accidental carry-over misses before XML assembly.
    4. AI Audit phase: Review only low-confidence items and anomalies, citing specific pixel coordinates for human verification.
    5. Delivery phase: Render final assets via MuseScore CLI and produce a mandatory human-verification checklist.

    Frameworks & tools

    Python-based workflow utilizing Audiveris for OMR, MuseScore 4 for rendering, and custom scripts for geometric pixel analysis. Supports MusicXML, PDF, and MP3 outputs.

    Why this beats prompting it yourself

    Generic prompts often lead to "token-burn" where the AI attempts to describe every pixel in ASCII, costing a fortune and hallucinating note pitches. This skill offloads the heavy lifting to specialized code gates, using the LLM only for high-level arbitration and complex engraving style decisions.

    Use cases

    • Digitizing hand-written or vintage scores into editable MusicXML formats.
    • Auditing existing digital transcriptions for accidental carry-over and rhythmic errors.
    • Converting poor-quality scans into professional-grade vector PDFs for publishing.
    • Creating reference audio tracks for rare solo instrument arrangements.

    Known limitations

    Requires a minimum staff-line spacing of 8px; content below this threshold is deemed unusable. Handwriting overlays require manual human confirmation. Vector PDF output depends on MuseScore 4 environment availability.

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

    Requires source resolution >150 DPI (8px staff spacing). Non-ASCII file paths can cause MuseScore CLI failures. Handwriting requires human arbitration.

    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

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 6 days ago

    • Free with an Agensi account

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Environment Variables
    Listed10 days ago
    Updated6 days ago

    What's inside

    Frequently Asked Questions

    This skill uses specialized machine-drafting and code-based validation gates to handle the heavy lifting, reserving AI only for high-level engraving audits. This approach eliminates the 'token-burn' and hallucinations common when asking generic LLMs to interpret complex musical notation.

    The workflow outputs industry-standard MusicXML for editing in software like Sibelius or Finale, print-quality vector PDFs, and reference MP3 files. It supports various input formats, provided the scan meets the minimum 8px staff-line spacing requirement.

    You will receive the complete Python-based workflow, including the source auditing scripts, validation gates, and engraving rule enforcement modules. It also includes the necessary configuration files to integrate with Audiveris and MuseScore 4.

    This skill is designed for Python environments and requires Audiveris for OMR and MuseScore 4 for rendering. It can be integrated into most AI agent frameworks that support local script execution and file system access.

    The 'Automated Validation' phase executes scripted gates that check for rhythmic integrity, accidental carry-over, and barline consistency. If a conflict is found, the system uses music theory logic to flag the specific pixel coordinates for human review.

    New in Agents & Orchestration

    Free