Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ai suplex 777 execution cycle

    1

    The 7-week execution loop: plan the cycle, run the day as a bounded session, close it in writing, and let only evidence-backed lessons into memory. Every session starts smarter than the last.

    Free

    Security scanned
    ai suplex 777 execution cycle

    ai suplex 777 execution cycle

    Example session with this skill installed

    I'm starting a new session for the 'API-Gateway-Refactor' project. My energy is at a 4. I need to fix the auth middleware and update the README. The cut line is the middleware fix.

    • Read your context and instructions
    • Compiled the ai suplex 777

    I have opened the session contract for today's work on the API-Gateway-Refactor. The context core has been loaded, confirming the cycle goal of shipping the V2 spec by Friday.

    Session — 2023-10-24

    Mission: Resolve authentication bypass in middleware and document breaking changes.
    Tasks:

    • task-001: Refactor auth.ts to use new JWT validation logic.
    • task-002: Update README.md with V2 authentication headers.
    • task-003: Verify middleware with local test suite.
      Success metric: Local integration tests pass with 401 Unauthorized for expired tokens.
      Energy: 4/5
      Cut line: task-001 (Middleware refactor)

    Session Close

    1. What shipped?
      • Refactored src/middleware/auth.ts (Artifact: commit-7a2b9f).
      • Updated README.md (Artifact: docs/v2-auth.md).
    2. What broke?
      • The test runner initially failed due to a missing .env variable for the new public key.
    3. What did I learn?
      • Public key rotation requires a grace period in the config file to prevent immediate downtime.
    4. What is next?
      • task-004: Deploy middleware to the staging environment.

    Next steps

    • Verify the auth.ts changes against the staging environment config.
    • Promote the "Public key grace period" lesson to the procedural memory folder after two more successful deployments.

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

    About this skill

    Stop losing ground to context drift and forgotten decisions. This skill implements a rigorous 7-week execution cycle designed for long-running projects where plans tend to evaporate by mid-week. It forces agents to operate within bounded daily sessions, ensuring every work block starts with context and ends with a written artifact. By strictly controlling how lessons are promoted from daily observations to permanent memory, it prevents the agent from re-litigating settled questions or following unverified patterns.

    What it does

    • Enforces bounded sessions by requiring a contract-style opening and a structured written close for every work block.
    • Validates evidence-based memory by only promoting lessons to long-term storage after three successful repetitions.
    • Eliminates status ambiguity by banning vague states like "ready" in favor of terminal states like SHIPPED, LIVE, or BLOCKED.
    • Defines weekly rituals that force plan adjustments based on actual performance rather than intent.
    • Generates immutable artifacts for every session to ensure a verifiable audit trail of what was actually shipped.

    How it works

    1. Initialize the Cycle by defining a one-sentence goal and five weekly priorities with dated terminal states.
    2. Open the Session using a five-line contract specifying the mission, tasks, success metrics, and a mandatory cut line for high-risk days.
    3. Execute and Close by answering four specific questions regarding artifacts, failures, lessons, and next steps.
    4. Promote Knowledge through the memory loop, moving observations into procedural or semantic storage only after evidence-backed validation.

    Frameworks & tools

    This skill is platform-agnostic and relies on structured Markdown files and directory hierarchies. It works in any environment that supports file-based context, such as Claude Code, Cursor, or custom agent runtimes using RAG.

    Why this beats prompting it yourself

    Standard prompting often fails to maintain discipline over long timelines, leading to "done-ish" tasks and memory bloat. This skill provides a hard governance layer that treats every day as a bounded contract, preventing the agent from wandering off-track.

    Use cases

    • Engineering projects where complex decisions are frequently forgotten across sessions.
    • Multi-week development sprints that suffer from "Monday morning resets."
    • Auditable workflows where every action must be backed by a physical artifact.
    • Research tasks requiring strict validation before a hypothesis is accepted as truth.

    Known limitations

    This is not for short, one-off tasks under two hours. It requires consistent file-based record keeping to function. Over-planning can occur if the rituals are not kept under 15 minutes.

    Not for tasks under 2 hours. Requires consistent file-based records. Rituals must be timed to avoid overhead. No automated task management integration.

    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 13 days ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed13 days ago

    What's inside

    Frequently Asked Questions

    This skill is designed to prevent 'context drift' and memory loss in long-term projects by enforcing a structured 7-week cycle with mandatory session contracts and evidence-based memory promotion. It ensures your agent doesn't re-litigate decisions or lose focus mid-week by treating every work block as a bounded, auditable event.

    It is platform-agnostic and works with any environment that supports file-based context and Markdown, including Claude Code, Cursor, Windsurf, and custom RAG-based agent runtimes. As long as your agent can read and write to a directory, it can execute this workflow.

    You will receive the core logic for the execution cycle, a set of standardized Markdown templates for sessions and memory promotion, and a deployment guide to initialize the 7-week folder hierarchy. These components provide the 'governance layer' that forces the agent to adhere to the disciplined workflow.

    Unlike standard prompting, this skill uses a 'hard memory loop' that only promotes lessons to long-term storage after three successful repetitions. This prevents your agent's context from becoming bloated with unverified assumptions or one-off failures, keeping its knowledge base lean and accurate.

    Setup is straightforward: you initialize the directory structure using the provided templates and define your 7-week goal. Once the folders are in place, you simply point your agent to the 'Session Contract' template to begin the first work block.

    New in Agents & Orchestration