Menu
End-of-session workflow for shipping changes, consolidating memory, applying self-improvements, and preparing publishable outputs with safety gates.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Guideline for directly editing 3D-printing project files (.3mf), verified against BambuStudio specifically: cutting or otherwise modifying mesh geometry, wiring objects into the 3MF XML correctly, arranging multiple print plates in BambuStudio's shared multi-plate coordinate system, and setting print/project parameters. Use whenever a user asks to edit, cut, split, merge, reposition, rearrange, or reconfigure a BambuStudio .3mf project file by hand instead of through the slicer GUI. Other slicers (OrcaSlicer, PrusaSlicer, etc.) use similar-looking .3mf files but are NOT verified to follow the same plate-offset math or config schema — treat this skill's multi-plate guidance as BambuStudio-specific.
Upgrade Dokploy on a VPS running Docker Swarm. Covers version check, breaking-change review, pre-flight health checks, the official install.sh update path, host-mode port conflict handling, and post-upgrade verification. Use when the user asks to upgrade/update Dokploy on any fleet host.
Final pre-launch audit for a multilingual marketing landing page — SEO, AEO/AIO (structured data for AI search), GEO, and EU compliance (GDPR/ePrivacy). Catches the common failures that ship to production when a landing page is built on top of a shared app layer.
Triage and route anything that comes up during a session — bugs, ideas, improvements, small fixes — to the right place. Use when the user or you notice something that may not belong to the current task.
Rewrite English AI-generated or AI-polished drafts into natural, credible, human text. Use when the user asks to de-AI, remove AI tells, humanize English writing, make a draft sound human / less like ChatGPT, kill the AI smell, or fix tweets, replies, threads, LinkedIn posts, landing-page copy, cold emails, newsletters, blog posts, essays, bios, or marketing copy that reads machine-written. Improves authenticity and editorial quality. Do NOT use for pure translation, pure summarization, or factual research alone. Never promise to bypass AI detectors, fabricate facts/sources/experiences, or optimize for detector scores.
Attach to the user's OWN already-logged-in system browser over the Chrome DevTools Protocol (CDP) with Playwright, so automation reuses their existing session and avoids re-logging-in with 2FA. Use this ONLY when a fresh/headless browser won't do because the task needs the user's live login — e.g. 'do this in my browser', 'use my existing session/login', 'I'm already signed in, don't make me log in again', driving a console/dashboard behind 2FA while they watch. Do NOT use for ordinary browsing, scraping public pages, headless tests, or any automation that doesn't need their real session — use the agent's built-in browser tool / Playwright or chrome-devtools MCP / computer-use for those. Covers launching Chrome/Edge/Brave with a remote-debugging port, per-step one-shot Playwright scripts, screenshot-audited steps, shadow-DOM/overlay handling, handing control back at Touch ID/security-key/QR walls, one-time-secret capture, and cleanup. Own browser, own accounts, never to bypass auth.
| name | close-loop |
| description | End-of-session workflow for shipping changes, consolidating memory, applying self-improvements, and preparing publishable outputs with safety gates. |
| license | MIT |
| metadata | {"version":"2.3.6","category":"session-memory","license":"MIT","tags":["session-memory","workflow","shipping","handoff"],"hermes":{"tags":["session-memory","workflow","shipping","handoff"]}} |
Use this skill when the user says "close loop", "close-loop", "wrap up the session", "close session", "end session", "end of day", or invokes /close-loop. This is the full end-of-session controller (ship + memory + self-improve + publish); for archiving a single Trellis task use finish-work instead.
Run four phases in order and return one consolidated inline report.
This skill is split into components for maintainability:
components/01-design-principles.mdcomponents/02-phase-1-ship-state.mdcomponents/03-phase-2-memory.mdcomponents/04-phase-3-4-and-output.mdFollow these components in order.
01-design-principles.md)02-phase-1-ship-state.md)03-phase-2-memory.md)04-phase-3-4-and-output.md)Return:
references/memory-frameworks.mdassets/templates/wrap-report-template.md