A sophisticated persistent memory layer that enables AI agents to resume complex projects with full context of past conversations and decisions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install context-bridge
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-bridge 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 is designed to solve the fundamental problem of agent amnesia between sessions. While most AI interactions start from a blank slate, this skill creates a persistent bridge that saves conversation threads, architectural decisions, and file modifications. By integrating this into your Openclaw Skills workflow, you can return to a project after days or weeks and have the agent immediately recall the reasoning behind specific code changes and the current state of progress.
This skill goes beyond simple logging by reconstructing a semantic understanding of your work. It allows for a project-oriented approach where every session contributes to a growing institutional memory. Whether you are building complex software or managing long-term research, Context Bridge ensures that Openclaw Skills maintain the continuity required for professional development cycles.
Context Bridge Use Cases
- Resuming work on multi-day coding features without re-explaining requirements.
- Auditing project history to understand why specific technical decisions were made.
- Finding historical conversation threads related to specific bugs or features.
- Bridging context between different team members or long breaks in development.
- Generating project retrospectives and documentation from session histories.
How Context Bridge Works
- The skill monitors the active session to capture message history, command sequences, and file system changes.
- Key decision points and reasoning are extracted from the dialogue to create a high-level project summary.
- Data is serialized and stored locally in the Openclaw Skills workspace memory directory.
- Upon a resume command, the skill retrieves the relevant JSON data and injects the project history back into the agent's context window.
- Semantic indexing allows the user to query across multiple projects to find related sessions or specific pieces of information.
Context Bridge Setup
To get started with Context Bridge for Openclaw Skills, ensure you have jq and find installed on your system. Follow these steps to enable persistent memory:
- Enable the automatic save hook to prompt for context saving whenever you start a new session:
openclaw hooks enable context-bridge-autosave
- Save your first session to initialize the project storage:
echo "save context as my-feature-name" | openclaw agent
- Resume the project later using the resume command:
echo "resume context: my-feature-name" | openclaw agent
Context Bridge Data Schema & Taxonomy
Context Bridge organizes its memory within the Openclaw Skills workspace using a structured file hierarchy:
| Component | Path | Description |
|---|---|---|
| Project Data | contexts/[name].json |
Contains project metadata, summarized decisions, and tags. |
| Session Logs | contexts/[name]-sessions/ |
Individual JSON files for every discrete interaction session. |
| Search Index | index.json |
A consolidated index used for fast semantic search across all projects. |
The schema tracks specific fields such as decisions, files_changed, reasoning, and related_contexts to ensure a high-fidelity reconstruction of the project state.
Context Bridge Advanced Features
- Semantic Search: Find specific contexts using natural language queries rather than just exact project names.
- Decision Timelines: Export a chronological list of all project decisions into a Markdown file for documentation.
- Context Templates: Define standard structures for recurring tasks like bug fixes or feature development.
- Multi-Session Merging: Combine fragmented sessions from different days into a single, unified project context.
- Redaction Filters: Automatically scrub sensitive data like API keys and passwords before saving context to disk.
- Local Encryption: Secure sensitive Openclaw Skills project data with password protection.
SKILL.md
Loading
Related Openclaw Skills
Oblien Workspace Runtime
A secure, isolated Firecracker microVM environment providing a robust Internal API for AI agents to manage files, execute commands, and run interactive terminals.
Model Matrix
A weighted, policy-aware routing system for selecting the optimal AI model based on real-world performance, cost, and community sentiment.
Zettel Link
A semantic search and connection discovery tool for Obsidian vaults using local or cloud-based AI embeddings.
Weibo Timeline Monitor
Automates the monitoring and archiving of the Weibo Latest Posts timeline into organized, deduplicated Markdown files.
Crypto Trading Assistant
A command-line powerhouse for real-time cryptocurrency market analysis, portfolio tracking, and automated price alerting.
Foreseek
Foreseek enables natural language trading on prediction markets by matching your beliefs to Kalshi contracts.
Featured*
