- Home
- Skills
- Agents & Orchestration
- ai suplex 777 execution cycle
Works with the AI Agents you already use
ai suplex 777 execution cycle
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
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: Refactorauth.tsto use new JWT validation logic.task-002: UpdateREADME.mdwith 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
- What shipped?
- Refactored
src/middleware/auth.ts(Artifact:commit-7a2b9f). - Updated
README.md(Artifact:docs/v2-auth.md).
- Refactored
- What broke?
- The test runner initially failed due to a missing
.envvariable for the new public key.
- The test runner initially failed due to a missing
- What did I learn?
- Public key rotation requires a grace period in the config file to prevent immediate downtime.
- What is next?
task-004: Deploy middleware to the staging environment.
Next steps
- Verify the
auth.tschanges 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
- Initialize the Cycle by defining a one-sentence goal and five weekly priorities with dated terminal states.
- Open the Session using a five-line contract specifying the mission, tasks, success metrics, and a mandatory cut line for high-risk days.
- Execute and Close by answering four specific questions regarding artifacts, failures, lessons, and next steps.
- 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
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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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