Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    StreamTeem Project Builder Pipeline

    1

    It writes a high-level design with architecture, sequence, and data diagrams, then stops and waits for your approval.

    $1.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    StreamTeem Project Builder Pipeline

    StreamTeem Project Builder Pipeline

    Example session with this skill installed

    start a new work item called auth-service-migration and parse the requirements.txt file in the root.

    • Read your context and instructions
    • Compiled the streamteem project builder

    I have initialized the work item and processed the requirements. The pipeline is now at the requirements summary stage.

    Work Item: auth-service-migration

    PhaseStatusArtifacts Created
    ContextActiveproject-builder/active-context.txt
    Spec ParserCompletedocs/requirements-summary.md
    JournalUpdatedjournals/expert/progress.md

    Requirements Summary Highlights

    • Goal: Migrate session-based auth to JWT-based OIDC provider.
    • Constraints: Must support existing user IDs; zero downtime required during cutover.
    • Blocking Questions:
      • Which OIDC provider (Auth0, Okta, or self-hosted) is the target?
      • Are the legacy session secrets available for token signing during the transition?
    • Next Action: run hld-writer

    Next steps

    1. Review the docs/requirements-summary.md file for accuracy.
    2. Answer the blocking questions regarding the OIDC provider.
    3. Type continue to start the High-Level Design (HLD) phase.

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

    About this skill

    Project Builder Pipeline makes your agent work the way a senior developer or architect does: one step at a time, with the decisions written down.

    Say "start". The pack asks what you are building, turns your spec (or a description in chat) into a requirements summary, and raises the questions that would otherwise surface halfway through the code. It writes a high-level design with architecture, sequence, and data diagrams, then stops and waits for your approval. The detailed design follows, with a second approval gate. Only then does it break the work into numbered tasks, set up the environment, scaffold the project, and implement task by task with tests. The full test suite runs before anything is pushed.

    Everything the pack knows about the work lives in a project-builder/ folder in your repository: a journal with the next step, the design documents, the diagrams, the task list, and a memory of facts and rules about your project. Close the session, switch agents, come back next week, and "continue" picks up at the exact step where the work stopped.

    What makes it different:

    • "Continue" is never taken as approval. A design gate clears only when you approve the design.
    • Blocking questions block. The agent does not guess past a missing requirement.
    • It remembers. Corrections you give become rules that every later session follows.
    • After the first delivery, the same pack handles new features, changes, refactoring, and production incidents.

    Works with any agent that can read and write files and run commands. Needs git for the branch and pull request steps; a project without a remote works too.

    Requires a project-builder folder structure. Python 3 required for memory optimization scripts. Mermaid diagrams require a browser to view the generated HTML page.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    Requires

    Git
    Writable project folder
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access

    Shell runs git, tests, and setup commands. Read/Write create journals, designs, diagrams, and code under project-builder/ and the app. Network is used when pushing or opening a pull request.

    Listed4 days ago

    What's inside

    Frequently Asked Questions

    This pipeline solves 'context drift' by using a structured filesystem and persistent memory to track decisions, ensuring the agent doesn't forget project rules or design requirements halfway through implementation.

    The skill requires a specific folder structure to maintain its state and needs a Python 3 environment to run the memory analysis scripts that optimize long-term learning.

    Yes, while it includes a standard SDLC workflow, you can define your own developer roles, tech stacks, and decision-making rights during the initialization phase to suit your team's needs.

    Unlike standard prompting, this skill automatically updates a project journal and learned-facts database, preventing 'hallucinations' caused by context window limits in large codebases.

    You receive the complete pipeline logic, the memory management scripts, and the templates for the design gates, journal tracking, and Mermaid diagram synchronization.

    The pipeline is designed to be agent-agnostic, provided the agent has capabilities for file system access and executing command-line scripts for Git and Python.

    New in Agents & Orchestration