inception

    1

    Audit and disrupt AI-like patterns in text by replacing formulaic structures with intent-driven narrative decisions.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    inception

    Example session with this skill installed

    Perform a Quick Review on this project proposal. It feels a bit dry and repetitive, like it was written by a generic assistant. I need it to sound authoritative for the board.

    • Read your context and instructions
    • Compiled the inception

    I have completed a Quick Review of the project proposal using the document-text adapter. I identified two primary feature clusters where model-default convergence is weakening the authority of the text.

    Decision Ledger

    Feature ClusterConvergence EffectIntent RelevanceRecommendation
    Over-signaled TransitionsUses "Furthermore," "In addition," and "Consequently" to bridge every paragraph.Board expects concise, logic-driven flow without rhetorical hand-holding.Pattern-Disrupting Option: Remove transitional adverbs and rely on header-driven logic or direct thematic links.
    Hedged ClaimsFrequent use of "it appears that," "may potentially," and "it is suggested."Dilutes the requested "authoritative" tone and weakens the business case.Structural Option: Replace passive hedging with direct evidence-based assertions (e.g., "Data indicates" vs "It appears").
    Balanced TripletsLists risks and benefits in groups of three with identical rhythmic length.Creates a predictable, "sing-song" cadence typical of default model outputs.Keep Option: Retain the grouping if the three points are exhaustive; otherwise, vary length to highlight the primary risk.

    Next steps

    1. Approve the removal of transitional adverbs to tighten the logical flow.
    2. Identify specific data points to replace the hedged claims in the ROI section.
    3. Confirm if you want a Regression Check after these structural edits are applied.

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

    About this skill

    Generic, formulaic, and over-explained AI content can alienate readers and weaken professional arguments. This skill identifies model-default patterns in fiction, presentations, and documents to restore original intent and structural variety.

    What it does

    • Pattern deconvergence identifies clusters of AI-like cues that make text feel templated or formulaic.
    • Intent preservation maps every finding to your original goals to ensure structural changes don't distort your message.
    • Structural auditing diagnoses content decisions, like over-explaining or weak claim strength, rather than just swapping words.
    • Domain adaptation applies specific narrative, document, or social copy protocols to ensure context-aware feedback.
    • Regression checking verifies that requested changes haven't introduced new default patterns or broken intent constraints.

    How it works

    1. Selection: The agent chooses Quick Review for excerpts or Deep Audit for high-stakes, long-form content.
    2. Analysis: It identifies feature clusters, such as repetitive sentence structures or over-signaled transitions.
    3. Disposition: You receive a Decision Ledger or cards detailing the convergence effect and pattern-disrupting alternatives.
    4. Validation: After you approve changes, the agent runs a regression check or a JSON-validated audit script to ensure consistency.

    Frameworks & tools

    The skill utilizes the StoryScope feature registry for narrative inspection and includes a Python validation script for ledger integrity. It operates through specific adapters for fiction, presentation decks, documents, and social copy.

    Why this beats prompting it yourself

    Standard prompts like "make this sound human" usually lead to forced informality or generic word swaps. This skill uses a formal evidence-to-cluster trace to fix the structural reasons why text feels robotic, preserving your domain constraints and authority.

    Use cases

    • Fiction polishing to remove narrative hand-holding and generic scene beats.
    • Executive presentations to strip away over-explained transitions and formulaic slide copy.
    • High-stakes documentation to ensure policy or technical specs don't sound like AI-generated templates.
    • Social copy auditing to identify organic brand voices buried under model-default phrasing.

    Known limitations

    This is not an authorship detector or a visual design tool. It does not provide advice on typography, layout, or ad targeting. It will not auto-rewrite material decisions without approval.

    Not for authorship detection or cosmetic humanization. No support for visual design, layout, or paid social ad mechanics. Requires human disposition for all material revisions.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 27 days ago

    • Free with an Agensi account

    Requires

    Python
    Git

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    While simple prompts often result in forced slang or 'casual' fluff, this skill uses a formal evidence-to-cluster trace to diagnose the structural reasons text feels robotic, such as over-signaled transitions and repetitive sentence rhythms.

    The purchase includes the core Inception logic, the StoryScope feature registry for narrative auditing, a Python validation script for ledger integrity, and four specialized adapters for fiction, decks, documents, and social copy.

    Inception is designed to work with most major LLMs that support function calling or JSON outputs, provided they have sufficient context windows to process long-form Deep Audits.

    No, the skill generates a 'Decision Ledger' or advisory cards; it identifies pattern-disrupting alternatives but ensures the user retains final creative control over all material narrative decisions.

    By neutralizing 'model-default' phrasing and formulaic structures, the skill ensures that policy documents, technical specs, and executive slides maintain professional authority and avoid looking like generic templates.

    Inception performs a 'Regression Check' after changes are made to verify that the edits haven't inadvertently introduced new AI-like patterns or compromised your original intent.

    New in Technical Documentation