Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Quick Quote: Defensible Hours for One Feature

    1

    - **Splits the feature into 3 to 8 tasks**, with testing and review as tasks of their own.

    Free

    Security scanned
    Quick Quote: Defensible Hours for One Feature

    Quick Quote: Defensible Hours for One Feature

    Example session with this skill installed

    Quote password reset by email for our Next.js SaaS. Auth and Postmark already exist. I charge $60 an hour.

    • Read your context and instructions
    • Compiled the quick quote

    Example run.

    Quick Quote: Password reset by email for a Next.js SaaS

    TaskOptimisticLikelyPessimisticExpected
    Reset request form and email-sending endpoint1.52.542.58
    Signed, expiring reset token and storage123.52.08
    New-password page with validation11.531.67
    Email template and provider wiring0.5121.08
    Tests: token expiry, reuse, unknown email1232
    Review, fixes and staging check11.52.51.58
    Total11

    Expected = (optimistic + 4 x likely + pessimistic) / 6. Rows are rounded to 2 decimals; the total is summed before rounding.

    Quote: 11 to 12.25 hours (expected 11 h, plus 10% contingency for medium confidence; rounded up to the quarter hour).

    Cost at USD 60/h: USD 660.00 to USD 735.00.

    Assumptions

    • Auth and a transactional email provider already exist
    • One locale, no SMS reset

    Not included

    • Design work beyond the existing component library
    • Production deployment window and monitoring

    Hours are active working hours, not calendar time.

    Client paragraph

    Password reset by email will take 11 to 12.25 hours of work, which is USD 660 to 735 at your rate. That covers the request form, a signed token that expires, the new-password page, the email template, tests for expiry and token reuse, and a review on staging. It assumes your current auth and email provider stay as they are.

    What would move it most: a second locale, or an SMS option, would each add a task.

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

    About this skill

    A client asks how long one feature will take and you need an answer today. Guess low and you work the extra hours for free. Pad it and you lose the job, and either way you can't show how you got the number.

    What your agent does

    • Asks at most 3 questions, and only ones that change the hours, like whether auth or an email provider already exists.
    • Splits the feature into 3 to 8 tasks, with testing and review as tasks of their own.
    • Checks the usual forgotten work: migrations, error states, config, tests, review fixes and a staging check.
    • Gives each task three numbers (optimistic, likely, pessimistic) in active working hours.
    • Runs a small calculator that weights each task (O + 4M + P) / 6, adds 5, 10 or 15% contingency by confidence and rounds up to the quarter hour.
    • Writes a short client paragraph with the range, the assumptions and what is excluded.

    Built-in guard rails

    • Numbers come from the script, never retyped or adjusted by hand.
    • No invented rates: cost appears only when you give an hourly rate.
    • Quotes are ranges in working hours, with no delivery date attached.
    • More than 12 tasks? The skill treats it as a project and stops.

    What's inside

    SKILL.md, 1 Python script (standard library only, no network calls), 1 worked example and 25 automated tests.

    Need a whole project with overheads and an AI versus manual comparison? That is what AI-Era Software Estimator on Agensi is for.

    Works with Claude Code, Codex, Cursor and other agents that load SKILL.md.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Product Management.

    It estimates one feature at a time; anything over 12 tasks is treated as a project. The hours are only as good as the task breakdown, so read the task list before you send a quote. It does not track actual time against the estimate.

    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 today

    • Free with an Agensi account

    Software
    Claude Code
    orCodex
    orCursor
    Python 3.8 or newer
    Optional
    Your project repository

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listedtoday

    What's inside

    Frequently Asked Questions

    It gives a range in working hours, plus cost only if you share your hourly rate. You decide what to quote.

    No keys at all. If your repo is open in the agent it reads the code the feature touches, otherwise it works from your description.

    An agent that loads SKILL.md and Python 3.8 or newer for the calculator.

    Leave the rate out. The hour range still tells you whether your fixed price is safe.

    New in Product Management

    Free