Execute qualquer Skill no Manus
com um clique

Execute qualquer Skill no Manus com um clique

main-chat-handoff

Estrelas20

Forks0

Atualizado14 de junho de 2026 às 14:30

Turn a completed side-chat discussion into one paste-ready prompt for the main chat. Use when Igor asks for a prompt, handoff, copy-paste instruction, or main-thread message after using a side conversation to debate, inspect, or decide implementation details.

Instalação

Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.

Executar no Manus

Fonte

igor9silva

igor9silva/meseeks

Abrir repositório GitHub Ver repositórios do creator

Download

Executar no Manus

Ocupações relacionadasSOC

Baseado na classificação ocupacional SOC

Desenvolvedores de softwareInformática e Matemática·SOC 15-1252

Explorador de arquivos

2 arquivos

SKILL.md

readonly

name	main-chat-handoff
description	Turn a completed side-chat discussion into one paste-ready prompt for the main chat. Use when Igor asks for a prompt, handoff, copy-paste instruction, or main-thread message after using a side conversation to debate, inspect, or decide implementation details.

Main Chat Handoff

Produce the exact text Igor can paste into the main chat. The main chat has no side-chat context, so the prompt must be self-contained, settled, and actionable.

Workflow

Read the recent side-chat discussion and identify the final settled instructions.
Separate decisions from exploration. Keep decisions; drop discarded options, debate history, and side-chat provenance.
Convert conclusions into direct imperatives for the main chat.
Make the prompt contextless: include names, paths, API boundaries, invariants, and acceptance criteria needed to act without knowing this side chat existed.
Output only the prompt text. Do not introduce it with "use this", "paste this", or similar wrapper text.

Rules

Do not ask the main chat to redo research already completed in the side chat.
Do not say "likely", "probably", "maybe", "if needed", or similar hedges for settled conclusions.
Do not mention experimental branches, previous checks, or side-chat evidence unless Igor explicitly wants that in the prompt.
Do not tell the main chat to "check X for reference" when the side chat already checked X and reached a decision.
Do not include implementation search commands unless Igor asks for verification commands in the prompt.
Do not wrap the whole prompt in a Markdown code block. Do not use any code blocks.
Do not repeat the same rule in multiple sections.
Do not include verification commands; this is main chat's concern.
Preserve Igor's exact requested names, paths, statuses, route names, file paths, literals, and vocabulary.
If something is unresolved, either ask Igor before producing the handoff or mark it explicitly as a question for the main chat. Do not smuggle uncertainty into requirements. Assume it'll be blindly pasted.

Mais deste repositório

mesmo repositório

ticktick-import

igor9silva/meseeks

Import tasks from TickTick into the Meseeks task system, sync link groups from TickTick Inbox, or create TickTick tasks from local link files. Use when the user asks to fetch TickTick tasks, import from TickTick, sync TickTick, pull tasks from TickTick, discover their inbox ID, or push curated local links back into TickTick.

2026-07-0420

create-task

igor9silva/meseeks

Create simple Meseeks task or reference folders in the right public/private root. Use when the user asks to create a task, backlog item, inbox capture, reference note, imported task, or source-backed task file.

2026-07-0420

scrape-content

igor9silva/meseeks

Scrape tweets and web content, saving results as structured markdown files. Use when the user asks to scrape a tweet, scrape a URL, fetch tweet content, save a link, archive a post, or process a list of links to scrape.

2026-07-0420

unclutter

igor9silva/meseeks

Unclutter the Meseeks task system by directly organizing messy task folders, references, imports, tags, status tags, duplicates, stale active work, and public/private placement. Use when the user asks to unclutter, clean up tasks, organize everything, or fix task-system clutter.

2026-06-0520

learn-from-task

igor9silva/meseeks

Read a full Meseeks task conversation from Convex MCP, especially when the user says Learn from task, asks to read a /task/ URL, references a PROD task, or wants project instructions improved from a prior task failure. Use this to recover the original user intent, assistant visible say() output, internal action/debug timeline, and steering corrections, then update MasterPlan, skills, or task files so the mistake does not repeat.

2026-05-1820

learn

igor9silva/meseeks

Extract lessons from conversation mistakes and update project rules, skills, and relevant tasks to prevent recurrence. Use when the user says "learn" or "/learn", asks to update rules from mistakes, or wants to capture lessons from a completed task.

2026-05-1820

name	main-chat-handoff
description	Turn a completed side-chat discussion into one paste-ready prompt for the main chat. Use when Igor asks for a prompt, handoff, copy-paste instruction, or main-thread message after using a side conversation to debate, inspect, or decide implementation details.

Main Chat Handoff

Produce the exact text Igor can paste into the main chat. The main chat has no side-chat context, so the prompt must be self-contained, settled, and actionable.

Workflow

Read the recent side-chat discussion and identify the final settled instructions.
Separate decisions from exploration. Keep decisions; drop discarded options, debate history, and side-chat provenance.
Convert conclusions into direct imperatives for the main chat.
Make the prompt contextless: include names, paths, API boundaries, invariants, and acceptance criteria needed to act without knowing this side chat existed.
Output only the prompt text. Do not introduce it with "use this", "paste this", or similar wrapper text.

Rules

Do not ask the main chat to redo research already completed in the side chat.
Do not say "likely", "probably", "maybe", "if needed", or similar hedges for settled conclusions.
Do not mention experimental branches, previous checks, or side-chat evidence unless Igor explicitly wants that in the prompt.
Do not tell the main chat to "check X for reference" when the side chat already checked X and reached a decision.
Do not include implementation search commands unless Igor asks for verification commands in the prompt.
Do not wrap the whole prompt in a Markdown code block. Do not use any code blocks.
Do not repeat the same rule in multiple sections.
Do not include verification commands; this is main chat's concern.
Preserve Igor's exact requested names, paths, statuses, route names, file paths, literals, and vocabulary.
If something is unresolved, either ask Igor before producing the handoff or mark it explicitly as a question for the main chat. Do not smuggle uncertainty into requirements. Assume it'll be blindly pasted.