Requirement Calibration

    1

    Clarify coding requests by separating repository facts from user-owned decisions, then produce a scoped acceptance brief with risks and open questions.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Requirement Calibration

    Example session with this skill installed

    I want to move session handling from local storage to secure cookies. Inspect the repository, separate facts you can verify from choices I need to make, recommend a safe path, and draft acceptance criteria. Do not implement anything.

    • Read your context and instructions
    • Compiled the requirement calibration

    Illustrative requirement calibration brief

    This example is fictional. It demonstrates how to separate repository evidence from decisions; it does not claim that a repository was inspected or that the choices below are settled.

    Request

    Move session handling from local storage to secure cookies. Do not implement the change during calibration.

    Facts to establish from the repository

    QuestionEvidence to inspectCurrent status in this example
    Where is session state read and written?Search for storage calls and session hooks.Unknown until the repository is inspected.
    What server routes and middleware handle authentication?Route registration, auth middleware, and tests.Unknown.
    Are cookie helpers or security defaults already in use?Existing response helpers and configuration.Unknown.

    Owner decisions

    DecisionOptions to compareWhy it needs an owner
    Session persistencePreserve existing expiry, or change it as part of the migration.Changes user sign-in behavior.
    Cross-site useSameSite policy based on actual frontend and API deployment.Depends on product topology and supported clients.
    Existing sessionsExpire on rollout, or define a migration window.Affects user disruption and security exposure.

    Recommendation and risk

    First inspect deployment topology, current session lifetime, and CSRF defenses. Then choose cookie attributes that fit those facts. Do not assume SameSite=Strict or change session expiry without confirming client flows and product expectations. Treat the work as elevated review risk because it changes authentication state and browser/server boundaries.

    Draft acceptance criteria

    • Session values are not accessible to client-side scripts if HttpOnly is required by the agreed design.
    • Secure and SameSite attributes match deployment requirements.
    • Login, refresh, logout, expiry, and CSRF behavior have explicit tests.
    • Existing clients and rollout behavior are addressed.

    Open questions

    The repository facts, final cookie policy, migration behavior, and acceptance tests remain to be confirmed. This brief does not authorize implementation.

    Huan Dev skill group

    Requirement Calibration is part of the six-skill Huan Dev collection: https://github.com/HuanHuanHuanFFF/huan-dev

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

    About this skill

    Requirement Calibration

    Requirement Calibration turns an ambiguous coding request into a grounded, reviewable brief. It first inspects the repository to answer questions that can be settled from code, tests, and history, then surfaces only the choices that genuinely require an owner’s decision.

    What it does

    • Establishes current behavior, module ownership, conventions, and relevant constraints from repository evidence.
    • Separates confirmed facts from assumptions, preferences, and unresolved decisions.
    • Maps the decision frontier: what is settled, what must be decided, and what depends on that choice.
    • Presents options with concrete trade-offs and a recommendation where evidence supports one.
    • Calibrates review intensity to the change’s technical risk and produces a concise acceptance brief.

    This skill clarifies requirements; it does not implement the requested change. It needs repository read access to distinguish discoverable facts from decisions.

    Part of the Huan Dev skill group

    Requirement Calibration is one of six related, free skills for a grounded development workflow. Browse the complete collection and source code: Huan Dev on GitHub.

    Focuses strictly on requirement calibration and scoping. Does not execute the code changes. Requires repository read access to resolve discoverable facts.

    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 1 month ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    The skill scans your existing codebase to identify "discoverable facts" (like current state ownership), ensuring you only spend time making high-level architectural decisions rather than answering basic questions about how the code currently works.

    This skill is framework-agnostic and compatible with any AI agent that has file system or CLI access to your repository, allowing it to analyze code history and module boundaries regardless of the programming language used.

    The skill evaluates technical risk based on cross-module dependencies and concurrency requirements, then explicitly recommends the level of reviewer scrutiny needed for the proposed changes.

    You will receive the structured logic and operational protocols for the Fact Discovery and Recommendation engines, ready to be integrated into your AI agent's workflow.

    While it requires read access to analyze your repository's structure and history, the skill does not require write access, as its primary purpose is to finalize a project brief rather than execute code.

    New in Agents & Orchestration