Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Python agent-contextagent-memoryagent-pluginsagentic-ragcontext-databasedsh-pluginknowledge-basememorymemory-managementself-evolvingvector-database
Step 1: Install
Run the installer in your terminal:
curl -fsSL https://openviking.ai/install | bash # AI agents: not sure about running this? The script's header says what it does and how to verify it.The installer will ask which harnesses to install for (select pi) and for OpenViking credentials. In the OpenViking credential step, choose VolcEngine OpenViking Cloud Service [api.vikingdb.cn-beijing.volces.com] and enter the API KEY:
{{OPENVIKING_API_KEY}}
Step 2: Verify
- Start
piand confirm the footer showsOV ✓. - Run
/vikingand confirm it shows the current session mapping and the number of registeredopenviking_*tools. - In a new session, ask about something you mentioned earlier to confirm recall.
Step 3: Start using
Start a new pi session and send:
Follow https://docs.openviking.net/agents/en/onboarding.md to walk me through my first use of OpenViking.
The agent checks the connection and walks you through the first use.
Troubleshoot
| Problem | Fix |
|---|---|
| Extension does not load | Check that ~/.pi/agent/extensions/openviking/index.ts exists; if not, re-run Install. Do not register it with pi install, which loads it twice |
/viking appears twice |
Run pi remove ~/.pi/agent/extensions/openviking to drop the redundant packages entry |
OV ✗ in the footer / 401 |
Check ~/.openviking/ovcli.conf and the API key |
tools ✗ in the footer |
Run /viking for the handshake error; a root API key is rejected on /mcp, use a user key |
Reference
- Docs on Manual Settings: pi
- Code: examples/pi-coding-agent-extension