storyledger voice

    1

    Novelists often struggle with "samey" dialogue or characters who lose their unique rhythm halfway through a draft.

    $10

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    storyledger voice

    Example session with this skill installed

    Run a voice report on the /manuscript folder. Sarah and her mother seem to sound too similar in the later chapters. Check for drift and similar pairs.

    • Read your context and instructions
    • Compiled the storyledger voice

    I have completed the voice analysis for the manuscript. The attribution engine processed 84% of dialogue paragraphs. I found that Sarah and Mrs. Miller (her mother) have a fingerprint distance of 0.11, confirming they are linguistically similar, particularly in their use of hedges and average sentence length.

    Voice Analysis Report

    SpeakerLinesWords/LineFingerprintDistinctive Words
    Sarah2428.20.42actually, maybe, listen
    Mrs. Miller1568.50.38dear, suppose, actually
    James984.10.76no, stop, enough

    Findings

    V-001: Similar Pair (Sarah & Mrs. Miller)

    • Status: Open

    Observation: Distance 0.11. Both characters use "actually" as a frequent hedge and share a median sentence length of 8 words.

    Evidence: Sarah (Ch 4, Line 112): "I actually don't know." vs Mrs. Miller (Ch 5, Line 45): "It actually doesn't matter."

    Recommendation: Consider giving Mrs. Miller more formal diction or removing her use of "actually" to differentiate the generations.

    V-002: Voice Drift (Sarah)

    • Status: Open
    • Observation: In Chapter 14, Sarah's fingerprint shifts to 0.18 (Baseline 0.42).
    • Evidence: Sentence length drops to 3.2 words/line across 12 lines.

    Note: If Sarah is under duress here, this drift is likely an intended arc. If not, the voice has become uncharacteristically blunt.

    V-003: Flat Voice (James)

    • Status: Open
    • Observation: James has high distinctiveness but very low vocabulary variance.
    • Recommendation: Expand his "tough guy" persona to include specific observational vocabulary.

    Next steps

    1. Review the similarity between Sarah and Mrs. Miller in Chapter 5.
    2. Confirm if Sarah's brevity in Chapter 14 is intentional for the scene's tension.
    3. Run again with --names "Mrs. Miller=Mom,Mother" to capture missing attributions.

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

    About this skill

    Novelists often struggle with "samey" dialogue or characters who lose their unique rhythm halfway through a draft. Identifying these shifts manually in a 80,000-word manuscript is prone to human error and bias.

    What it does

    • Dialogue attribution isolates quoted speech from narration using tags, beats, and alternation patterns.
    • Voice fingerprinting measures sentence length, word choice, contractions, and hedges for every speaker.
    • Similarity detection flags character pairs whose speech patterns are statistically too close, risking reader confusion.
    • Drift analysis identifies specific chapters where a character's voice deviates significantly from their established baseline.
    • Bible verification compares actual dialogue against stated character traits in your story bible to ensure consistency.

    How it works

    1. Scan manuscript by pointing the agent to your chapters/ or manuscript/ folder.
    2. Execute measurement via a Node-based script that generates fingerprints for each identified speaker.
    3. Review attribution to resolve aliases or roles (e.g., mapping "Mom" to "Sarah") using provided flags.
    4. Analyze report to identify flat voices, unintended similarities, or justified character arcs.

    Frameworks & tools

    This skill requires Node 18 or later and operates on standard .md or .txt manuscript files. It integrates with characters/*.yml files if a StoryLedger story bible is present.

    Why this beats prompting it yourself

    Generic prompts often hallucinate dialogue patterns or fail to track consistency across large file sets. This skill uses a specialized script to provide hard data on linguistic markers, ensuring your analysis is based on evidence rather than an LLM's vibe-check of a single chapter.

    Use cases

    • Ensure an ensemble cast of five distinct protagonists all sound unique.
    • Detect if a character's speech patterns change unintentionally after a long writing break.
    • Verify that a character described as "succinct" isn't actually using high word counts.
    • Identify every line of unattributed dialogue to tighten prose clarity.

    Known limitations

    Optimized for English punctuation patterns. Dialogue using non-standard quoting or specific dash-led formats may result in lower attribution rates. Requires "Chapter N" headings for drift analysis in single-file manuscripts.

    English only. Requires standard quotes/dashes for attribution. One-file drafts need Chapter headings for drift detection. Unattributed lines may be high in books with sparse tagging.

    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 any payment.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 5 days ago

    Permissions

    Terminal / Shell
    Read Files
    Network Access
    Write Files

    File Scopes

    chapters/**
    manuscript/**
    draft/**
    *.md
    *.txt
    characters/**
    voice/**
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    The skill uses linguistic markers like sentence length, hedge words, and contraction frequency to create a statistical 'fingerprint' for each character, allowing you to objectively see if your cast sounds too similar or if a protagonist’s voice changes over time.

    This skill requires Node 18+ and is designed for standard `.md` or `.txt` files; it also integrates seamlessly with StoryLedger-formatted story bibles (.yml) to compare actual dialogue against intended character traits.

    Unlike generic prompts that rely on an LLM’s 'vibe-check' and limited context window, this skill uses a specialized script to process your entire manuscript at once, providing hard data and consistent tracking across hundreds of pages.

    The package includes the core Node-based measurement script, the dialogue attribution logic, and the reporting module, along with documentation on how to map character aliases for accurate tracking.

    The tool is currently optimized for English punctuation (e.g., standard quotation marks); while it can process other languages, attribution accuracy may vary if the manuscript uses non-standard quoting or dash-led dialogue formats.

    New in Writing & Content