Synap is a sovereign AI knowledge layer providing structured persistent memory, typed entity graphs, and messaging relays powered by PostgreSQL.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install synap
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 synap 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 Synap?
Synap serves as a robust, sovereign AI knowledge infrastructure designed to give agents a long-term, structured memory. As a core component within the Openclaw Skills ecosystem, it utilizes PostgreSQL for data persistence, Typesense for full-text search, and pgvector for semantic search capabilities. This combination allows agents to maintain a complex, typed knowledge graph that transcends simple chat history, ensuring that data is organized, searchable, and governed according to user-defined rules.
By deploying a Synap pod, developers can ensure their AI agents have a dedicated workspace to store entities, documents, and atomic facts. The system is built on principles of data sovereignty, meaning the user retains control over their information while the AI gains the ability to recall context, manage tasks, and link related concepts through a sophisticated relationship model.
Synap Use Cases
- Establishing a long-term memory for AI agents to remember user preferences and past interactions.
- Building a structured knowledge base of contacts, projects, and tasks that persists across different sessions.
- Creating a centralized hub for cross-platform messaging by relaying Telegram, WhatsApp, or Slack communications.
- Automating document management and meeting summaries with a built-in governance and proposal system.
- Developing a complex entity-relationship graph to track business deals, company hierarchies, and professional networks.
How Synap Works
- The agent connects to a dedicated Synap pod using the provided API key and configuration URL.
- Information is captured and categorized into entities (like people or tasks), long-form documents, or atomic facts.
- Every write operation passes through a governance layer that either auto-approves the data or creates a proposal for user review.
- Entities are linked using typed relations to build a traversable knowledge graph.
- The agent performs semantic or keyword searches to recall relevant context based on current user prompts.
- External messages are ingested via threads, processed by the AI, and stored within the persistent knowledge base.
Synap Setup
To integrate Synap with Openclaw Skills, you can use the recommended automatic configuration by setting the following environment variables:
export SYNAP_HUB_API_KEY=hub_xxxx
export SYNAP_CONFIG_URL=https://pod.synap.live/trpc/intelligenceRegistry.getServiceConfig
For manual configuration, define the specific pod URL and workspace identifiers:
export SYNAP_HUB_API_KEY=hub_xxxx
export SYNAP_POD_URL=https://pod.synap.live
export SYNAP_WORKSPACE_ID=<uuid>
export SYNAP_AGENT_USER_ID=<uuid>
Synap Data Schema & Taxonomy
Synap uses a multi-layered data model to organize information within the Openclaw Skills framework:
| Data Type | Primary Function | Example Properties |
|---|---|---|
| Entities | Structured objects | status, priority, email, dueDate |
| Documents | Long-form Markdown | Meeting notes, project specs, reports |
| Facts | Atomic memory units | user preferences, specific data points |
| Relations | Graph connections | related, parent, child, belongs-to |
Standard profiles include note, task, project, event, person, contact, company, deal, bookmark, and article, all of which can be extended with custom properties.
Synap Advanced Features
- Multi-agent support via Agent-to-Agent (A2AI) communication threads.
- Hybrid search architecture combining Typesense full-text search and pgvector semantic similarity.
- Governance engine that handles automated approvals or manual user proposals for data integrity.
- Extensible profile system allowing for the creation of new entity types like leads or webinars.
- Comprehensive audit trails through mandatory reasoning fields for every write and update operation.
SKILL.md
Loading
Related Openclaw Skills
Free Text to Video Generator AI
A powerful AI video creation tool that converts written prompts, scripts, and documents into professional 1080p MP4 videos using cloud GPU rendering.
Text to Video Generator AI Free
A powerful AI-driven skill that converts written text, documents, and prompts into professional 1080p MP4 videos using cloud GPU rendering.
Image Video AI
A high-performance AI skill that converts static image collections into animated 1080p MP4 videos with transitions and background music.
Shorts Maker
Shorts Maker is a high-performance AI tool that transforms long-form video content into optimized vertical clips for social media using cloud-based GPU rendering.
Easy Video Editor
A high-performance AI video editing skill that automates trimming, background music integration, and 1080p rendering using cloud GPUs.
Best Avatar Video
A cloud-based AI agent skill that transforms static photos and text scripts into high-quality 1080p talking avatar videos.
Featured*
