agentsclimarketplace

Cross model handoff

Skill ntaffzii/Skill-Agents/Skill.md/cross-model-handoff

Personal AI-agent skills, workflows, prompts, and setup guides for local LLMs and MCP-powered assistants.

Install
npx -y skills add ntaffzii/Skill-Agents --skill cross-model-handoff

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Create continuation notes that another model, provider, or agent can resume without relying on hidden context or exact tool names. Use when the user asks to hand off work between Claude, ChatGPT, Gemini, Codex, local agents, or another MCP-enabled assistant.

SKILL.md

1.6 KB, as published. Nobody here has run it

Cross-Model Handoff

Use this skill when work may continue in a different model or client.

Core Rule

Write the handoff so a different agent can continue without access to this conversation's hidden context.

Workflow

  1. Capture objective

    • State the user's goal.
    • Include current scope and non-goals.

  2. Capture state

    • Files changed or inspected.
    • Decisions made.
    • Commands, tests, or checks run.
    • Tool capabilities used.

  3. Separate facts from assumptions

    • Verified facts
    • Assumptions
    • Unknowns
    • Blockers

  4. Translate tools

    • Say capabilities, not only exact tool names.
    • Example: "Used file search and read tools" instead of only "called search_in_files".
    • Include exact local names only as optional implementation detail.

  5. Give next actions

    • First next step
    • Validation step
    • Risk to watch

Output Format

# Cross-Model Handoff

## Objective

## Current State

## Decisions

## Files And Artifacts

## Tool Capabilities Used

## Verification

## Assumptions

## Next Steps

## Risks

Rules

  • Do not reference hidden conversation context as required knowledge.
  • Do not assume the next agent has the same tools.
  • Do not imply tests passed unless they actually ran.
  • Include exact paths, URLs, commands, or artifacts when they matter.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.