Huan Dev

    1

    A free, six-skill workflow collection for grounded repository changes: calibrate requirements, execute from a goal, pressure-review risks, and deliver an evidence-based brief.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Huan Dev

    Example session with this skill installed

    Use the Huan Dev workflow to plan a change in an existing repository: add retry handling to a payment webhook. First inspect the current code and tests, identify decisions that need my input, then implement only after the scope is clear. Report what you verified and what remains unknown.

    • Read your context and instructions
    • Compiled the huan dev

    Example workflow output (illustrative)

    This is an example of the kind of evidence-based handoff the Huan Dev collection aims to produce. It does not claim that a repository was inspected or code was changed in this listing.

    Acceptance brief

    ItemCurrent decisionEvidence needed
    Retryable failuresTo be confirmed against the existing handler and provider contractHandler code, provider documentation, current tests
    Maximum attempts and backoffUser decision if not already specified by project policyExisting configuration and operational limits
    Duplicate event handlingPreserve or define idempotency behavior before implementationEvent storage and uniqueness constraints
    Terminal failuresDefine logging, alerting, and response behaviorCurrent observability and retry contract

    Execution route

    1. Inspect the webhook handler, related persistence code, tests, and repository guidance.
    2. Identify any unresolved product or operational choices and ask only for decisions that cannot be inferred safely.
    3. Implement the agreed behavior within the existing architecture.
    4. Run focused tests, then the relevant broader checks.
    5. Review concurrency, duplicate delivery, state transitions, and public API behavior.

    Handoff format

    The final report should name changed files, summarize behavior, list the exact checks that ran and their results, and call out checks that were not run. If implementation was blocked, explain the evidence and the next decision needed.

    Collection links

    The focused skills are requirement-calibration, execute-from-goal, pressure-review, delivery-brief, and prompt-entropy. Browse the free collection and source at https://github.com/HuanHuanHuanFFF/huan-dev.

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

    About this skill

    Huan Dev

    Huan Dev is a free collection of six related skills for carrying repository work from an unclear request to a reviewable handoff. The main orchestrator connects the focused skills: requirement-calibration, execute-from-goal, pressure-review, delivery-brief, and prompt-entropy.

    Collection

    • Huan Dev — orchestrate the full workflow
    • requirement-calibration — resolve scope, authority, and acceptance
    • execute-from-goal — implement a concrete goal with verification
    • pressure-review — inspect consequential changes for risks and gaps
    • delivery-brief — report evidence, limitations, and next steps
    • prompt-entropy — create clear prompts for delegated or reusable work

    All six skills are free. The ZIP includes the references needed to use this listing standalone. For the complete collection, updates, source files, and issue tracking, visit the Huan Dev GitHub repository. Licensed under MIT.

    Requires repository context and history. Simple changes bypass the full pressure-review loop. Subskills must be available in the references directory.

    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

    Unlike standard AI prompts that jump straight to coding, this skill enforces a structured lifecycle—including requirement calibration and pressure testing—to prevent regressions and ensure code matches your repository's specific architecture and history.

    This skill is language-agnostic and designed to work across any repository as long as the AI agent has access to the version control history and live file state.

    The skill excels at complex, cross-boundary changes; however, it is designed with an internal logic that bypasses intensive 'Pressure Review' cycles for simple, local, or low-risk edits to maintain developer speed.

    Every completed task concludes with an 'Evidence-Bounded Handoff,' which is a structured report detailing the changes made, verification results, and any remaining architectural uncertainties you should review.

    The skill requires an environment where your AI agent has read/write permissions for the repository and the ability to inspect Git history to ensure proper context grounding.

    New in Agents & Orchestration