model-handoff
Maintain a HANDOFF.md file in the workspace so context survives seamlessly when switching between LLM models (e.g. Claude → GPT → Gemini). Use when the user says they are switching models, asks how to preserve context across model switches, wants to save tokens by rotating models, or asks how a new model can pick up where the last one left off. Also use proactively during long sessions to keep HANDOFF.md current.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/bwiley1989/model-handoffWhat This Skill Does
The model-handoff skill acts as a crucial bridge for context retention across different LLM sessions. As users transition between various AI models like Claude, GPT, or Gemini, vital project information, behavioral constraints, and task status are often lost. This skill automates the creation and maintenance of a HANDOFF.md file in the workspace root. This file functions as a 'fast-boot' manifest, enabling any new model to instantly inherit the project state, active file references, and necessary behavioral rules. By centralizing this information, the agent ensures that no momentum is lost during a switch, effectively creating a persistent thread of memory for an otherwise ephemeral model session.
Installation
To integrate this skill into your environment, run the following command in your terminal:
clawhub install openclaw/skills/skills/bwiley1989/model-handoff
Once installed, ensure your AGENTS.md file is updated to include a directive for new models to read HANDOFF.md immediately upon initiation.
Use Cases
- Model Rotation: When switching models to optimize for cost, speed, or reasoning capabilities, use this skill to preserve your progress.
- Long-running Projects: During extended development sessions, use this to summarize complex state changes, ensuring a new model doesn't re-tread ground already covered.
- Handing Off Context: When you need a specialized model to take over a specific coding task, this skill ensures the new agent understands its identity and current project constraints.
- Proactive Maintenance: Keep documentation synchronized automatically so that you are never caught off-guard by a model reset.
Example Prompts
- "I'm switching to GPT-4o for this coding task; please update the HANDOFF.md file so the new model knows exactly where I am."
- "How can I make sure the next model I use doesn't lose the context of our current work on the API refactor?"
- "We've been working for hours—please generate a fresh HANDOFF.md with all our current project files and the rules you've been following."
Tips & Limitations
- Stay Dense: The
HANDOFF.mdshould be a summary, not a novel. Focus on file paths, current status, and critical constraints. - Security: Never place secrets, passwords, or personal keys directly into the
HANDOFF.mdfile. Always point to external credential files instead. - Timestamps: Always check the 'Last updated' section to ensure the context you are inheriting is still relevant to the current state of your code.
- Proactive Updates: Don't wait for a crash or a forced restart; trigger an update whenever you finish a major milestone or reach a stable breaking point in your project.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-bwiley1989-model-handoff": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-write, file-read
Related Skills
safe-long-run-mode-gpt54
Operate long-running tasks safely when the environment is optimized for GPT-5.4 as the primary and often only model. Use when the user wants a low-cost, high-throughput long-run workflow, plans to keep everything on GPT-5.4, or asks how to run long coding, research, build, documentation, Azure, or multi-agent tasks safely without relying on Claude.
safe-long-run-mode-gpt54-claude
Operate long-running AI tasks safely across GPT-5.4 and Claude by using model selection rules, phased execution, checkpoints, resumable workflows, API throttling discipline, and subagent isolation. Use when a task may run for a while, touch multiple files/systems, involve external APIs, browser automation, Azure, Orgo, or multiple subagents, or when the user asks about long autonomous runs, rate limits, reliability, or safe operating mode.