BEZEN Sentence Ladder

    1

    What stands behind a harsh sentence, and four small steps to the version people can hear, with the trait between its two extremes.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    BEZEN Sentence Ladder

    Example session with this skill installed

    Ladder this sentence my team lead said in our meeting: "You never finish anything."

    • Read your context and instructions
    • Compiled the bezen sentence ladder

    What stands behind it

    • Thinking pattern: a judgment of the whole person ("never", "anything")
    • Emotion: frustration and overload
    • Body: tight chest, voice rising
    • Behavior: blames the person instead of describing the report
    • Silent belief: "If I don't push, nothing will move."

    The ladder

    1. "You never finish anything."
    2. "It's not done again, and I'm about to explode."
    3. "Yesterday's report still isn't done, and it's blocking me."
    4. "The report isn't done and I need it by five. What's stuck, and how can I help you finish?"

    The trait

    Constructive criticism, between blaming and giving up.

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

    About this skill

    Free during launch, until 31 Jan 2027.

    In one sentence: Give BEZEN one harsh sentence someone said, and get two things: what stands behind it, and a ladder of four small steps to the version that can be heard.

    Who: Used by teachers and counselors, team leads, parents, group facilitators. The sentence comes from a child in class, an employee in a meeting, a partner at home, someone in a WhatsApp group.

    When: after a hurtful sentence was said, before a hard conversation you need to prepare, or when building a lesson on communication.

    Where: class, home, work, social media. The structure is the same everywhere; only the words change.

    What stands behind the sentence (the core of BEZEN): before the ladder, BEZEN reads the sentence in five lines:

    1. Thinking pattern, e.g. "never" and "anything", a judgment of the whole person.
    2. Emotion: what is burning right now.
    3. Body sensation: what the body is doing at this moment.
    4. Behavior: what the person does, not what they feel.
    5. Silent belief: the sentence under the sentence, the one that was not said.

    Anything the sentence does not show is written as "unknown". Nothing is invented.

    Why it works: people do not jump from attack to calm in one step. Each step is small, one or two percent from the one before, so someone standing on the previous step can really reach it. The hurtful behavior stops first; the emotion calms only after. The claim itself never disappears. And at the end the trait gets a name, always in the middle between two extremes.

    How: paste one sentence. You get a short page: the sentence, the five lines behind it, the four steps, and the trait with its two extremes. A code check makes sure the original claim is still there in every step.

    Full example

    The sentence: "You never finish anything."

    Behind it: a judgment of the whole person ("never", "anything") · frustration and overload · tight chest, voice rising · blames the person instead of describing the report · "If I don't push, nothing will move."

    • Step 1, as it was said: "You never finish anything."
    • Step 2, the blaming stops, the anger is still there: "It's not done again, and I'm about to explode."
    • Step 3, from the person to the problem: "Yesterday's report still isn't done, and it's blocking me."
    • Step 4, the full claim, with an opening to act together: "The report isn't done and I need it by five. What's stuck, and how can I help you finish?"

    The trait: constructive criticism, in the middle between blaming (the hard extreme) and giving up (the too-soft extreme).

    Before: Ask a regular AI to soften the sentence and you get: "Maybe we should try to finish things on time?" The claim is gone, and nobody knows what isn't done or by when.

    After: Step 4 above. The same claim, with the report, the time, and a question that opens a conversation.

    Known limitations:

    • One sentence at a time.
    • A learning and practice tool, not therapy.
    • Output is in Hebrew.

    Part of BEZEN (bezenai.com), an emotional-pattern layer for AI agents and bots.

    Supports Hebrew output only. Optimized for single sentences. Requires a local environment to run the verify.py script for full validation.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Listed4 days ago

    What's inside

    Frequently Asked Questions

    The skill uses a 4-step 'ladder' approach that preserves the original point of the message while stripping away destructive phrasing. Unlike generic prompts that produce overly soft results, this ensures the core critique or complaint remains intact and actionable.

    This skill is optimized for any AI agent that supports Python execution for its internal verification script. It is specifically designed to output results in Hebrew, making it ideal for Hebrew-speaking workplaces and educational environments.

    You receive the core logic for the 'Sentence Ladder' framework, the custom `verify.py` script for logical validation, and the structured workflow to integrate this into your existing AI agent or conflict resolution platform.

    The skill is designed to analyze single, impactful sentences. While you can input longer rants, the analysis is most effective when focused on the core toxic statement to ensure the psychological mapping remains accurate and the ladder stays reachable.

    Updates are periodically released to refine the behavioral mapping and the `verify.py` logic. Buyers have access to the latest version of the skill through their marketplace library, ensuring the 'Golden Path' traits remain relevant and effective.

    New in Productivity