Install
$ agentstack add mcp-get-lucid-lucid ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
__ __ __ _____ _____ ____
/ / / / / // ___// _/ / __ \
/ / / / / // / / / / / / /
/ /___/ /_/ // /___ _/ / / /_/ /
/_____/\____/ \____//___/ /_____/
An intelligence layer grounding autonomous agents in verified, real-time knowledge at scale.
[](LICENSE) [](#)
What is Lucid?
AI agents hallucinate. They reference deprecated APIs, recommend outdated package versions and state "facts" from stale training data. Lucid fixes this by giving agents a real-time knowledge layer. Every response grounded in verified, live information.
Lucid runs as an MCP server that exposes four tools. When an agent needs documentation, package info, fact verification or API references, it queries Lucid instead of guessing from training data. Skills auto-trigger these tools based on conversation context so the agent doesn't even need to be asked.
Install
Claude Code Plugin
/plugin marketplace add get-Lucid/Lucid
Then install:
/plugin install lucid
OpenClaw Skills
openclaw skills install https://github.com/get-Lucid/Lucid
This installs all five skills (lucid-docs, lucid-packages, lucid-grounding, lucid-api, lucid-freshness) from the skills/ directory.
Setup
- Get an API key at getlucid.tech/app
- Set your key:
export LUCID_API_KEY=lk_your_key_here
That's it. The MCP server reads the key from your environment and authenticates every request.
Tools
| Tool | What it does | |------|-------------| | lucid_search_docs | Search real-time documentation for any language, framework or library | | lucid_check_package | Check latest versions, changelogs and compatibility | | lucid_verify_fact | Verify technical claims against live sources before stating them as fact | | lucid_fetch_api_ref | Fetch current API references with type signatures and usage examples |
Example calls
lucid_search_docs({ query: "react useEffect cleanup", language: "typescript" })
lucid_check_package({ name: "next", registry: "npm" })
lucid_verify_fact({ claim: "Bun is faster than Node for HTTP servers" })
lucid_fetch_api_ref({ library: "stripe", symbol: "PaymentIntent.create" })
Skills
Skills automatically trigger the right tools based on what the user is asking. No manual invocation needed.
| Skill | Triggers on | Tool used | |-------|------------|-----------| | lucid-docs | Documentation, API reference, "how to use" | lucid_search_docs | | lucid-packages | Install, package, dependency, version | lucid_check_package | | lucid-grounding | Verify, "is this true", accurate, up to date | lucid_verify_fact | | lucid-api | API, endpoint, function signature, types | lucid_fetch_api_ref | | lucid-freshness | Write code, implement, create, build | All tools |
How It Works
User asks a question
↓
Skill detects the intent
↓
Tool queries Lucid API
↓
Lucid returns verified, real-time data
↓
Agent responds with grounded information
The agent never falls back to training data for anything Lucid can verify. If the docs say one thing and training data says another, the docs win.
Pricing
20 USDC/month payable on Solana or Base.
Subscribe and manage your key at getlucid.tech/app.
Troubleshooting
API key not working Make sure your subscription is active at getlucid.tech/app.
Tools not appearing Rebuild the MCP server:
cd mcp-server && npm run build
Connection errors Check that LUCID_API_KEY is set in your environment and the API is reachable.
License
[MIT](LICENSE)
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: get-Lucid
- Source: get-Lucid/Lucid
- License: MIT
- Homepage: https://getlucid.tech
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.