A unified continuity system that persists project context, skill registries, and session history across AI model switches and session resets.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install context-bridger
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install context-bridger using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Context Bridge?
Context Bridge serves as the essential continuity layer for the Openclaw ecosystem, specifically designed to solve the blank slate problem encountered during model transitions or session restarts. It acts as a bridge between isolated interactions, ensuring that your AI agent maintains a persistent understanding of its environment, including installed capabilities and ongoing projects. By utilizing this system, developers can switch between different LLMs or initiate new sessions without the need to manually re-explain project goals or technical configurations.
At its core, Context Bridge implements a robust architecture of registries and handoff memories that track Openclaw Skills, cron jobs, and agent configurations. This approach transforms the AI from a stateless tool into a context-aware assistant that recognizes previous actions and available resources automatically upon startup. It is an infrastructure-level enhancement that significantly improves the reliability and efficiency of multi-agent and multi-model workflows.
Context Bridge Use Cases
- Maintaining project continuity when switching between different LLMs like GPT-4 and Claude.
- Preventing context loss after using the /new command to clear session history.
- Managing a growing library of Openclaw Skills and ensuring they are discoverable by the agent.
- Documenting model-agnostic handoffs to keep track of work-in-progress tasks across different days.
- Centralizing the discovery of active cron jobs and scheduled automation scripts.
How Context Bridge Works
- The setup script performs a comprehensive scan of both local and system directories to identify all installed Openclaw Skills and agents.
- Registry files are generated in JSON format, creating a structured map of skills, crons, and discovery metadata.
- A session startup protocol is established via AGENTS.md, requiring the AI to read these registries at the beginning of every interaction.
- The model-agnostic memory file tracks the current state of projects and recent model actions, providing a chronological narrative for the next session.
- The AI agent acknowledges the loaded context in its initial response, confirming it has access to all registered tools and historical data.
- Any changes to the environment, such as adding new Openclaw Skills, are logged back into the registries to maintain an evergreen system state.
Context Bridge Setup
# 1. Install the skill
openclaw skills install https://github.com/loudmouthedmedia/context-bridge
# 2. Run the discovery and setup script
cd ~/.openclaw/workspace/skills/context-bridge/scripts
./setup.sh
# 3. Verify registries are created in ~/.openclaw/
Context Bridge Data Schema & Taxonomy
Context Bridge utilizes a structured file hierarchy to manage system state and memory across the following schema:
| Registry | Location | Purpose |
|---|---|---|
| Skills Registry | ~/.openclaw/skills-registry.json |
Tracks all installed Openclaw Skills and their source paths. |
| Cron Registry | ~/.openclaw/cron-registry.json |
Lists all active scheduled automation jobs. |
| Discovery Map | ~/.openclaw/skills-discovery.json |
Defines the specific capabilities provided by each skill. |
| Handoff Memory | ~/.openclaw/model-agnostic-memory/model-handoff.md |
Captures project status and actions for session continuity. |
| Agent Registry | ~/.openclaw/agents/*/agent.md |
Stores specific configurations for individual AI agents. |
Context Bridge Advanced Features
- Auto-discovery of local and system-wide Openclaw Skills via automated shell scripts.
- Model-agnostic handoff protocols that allow different LLM architectures to share a single source of truth.
- Strict file responsibility assignments (SOPs) to prevent technical instructions from leaking into personality profiles.
- Support for manual context reloading within active sessions using the load context command.
- Deep integration with AGENTS.md to enforce mandatory context-awareness on every session startup.
SKILL.md
Loading
Related Openclaw Skills
Stock Macro & Market Analysis Skill
A professional-grade market analysis tool that evaluates technicals, sentiment, and macroeconomics to generate comprehensive financial reports.
High-Intent Lead Sniper
A high-performance lead generation agent that detects real-time buying signals and generates hyper-personalized outreach for B2B sales teams.
AI PPT Generation
An intelligent presentation tool for Openclaw Skills that converts user themes and resource files into structured, professional PowerPoint decks.
KidsNews Daily Brief Skill
A specialized AI agent skill designed to deliver curated, educational news summaries for children and parents every morning.
Ultimate Lead Scraper and AI Outreach Engine
A powerful automated pipeline that scrapes B2B leads from multiple directories, scores them based on ICP fit, and generates 4-step AI outreach sequences.
Chinese Invention Patent Drafting Assistant
A specialized AI assistant that guides users through a multi-step dialogue to draft professional Chinese invention patent documents including claims and descriptions.
Featured*
