Use case
A web reader for agents that reads from the index.
Agents do not need a browser; they need the page, clean, now. Lyrenth gives any agent stack one reading surface into the AI-readable web index: an MCP tool, a one-line reader endpoint, and native SDKs, all returning the same stable AIDocument for every URL on the web.
Get API access →Free tier · 2,000 reads/mo · no credit card
MCP in one command
npx -y lyrenth-mcp adds read_url to Claude Desktop, Claude Code, Cursor, or any MCP client. Your agent reads the web through the index with zero glue code.
SDKs and adapters
pip install lyrenth or npm install lyrenth-sdk, with LangChain, LlamaIndex, and Vercel AI SDK adapters ready-made.
Context-budget friendly
Clean, low-token content with an economics block on every read, so agent loops stop burning context on markup and cookie banners.
Reliable under load
The cross-caller cache means your fleet of agents reading the same docs hits the index, not the origin: no rate-limit roulette mid-run.
Measured, not promised
Real pages, read through the production index, with the economics the API itself reports:
| GitHub REST API quickstart | 88,614 raw | 4,876 indexed | 94.5% saved |
| Kubernetes Pods concepts | 131,977 raw | 7,475 indexed | 94.3% saved |
| React useState reference | 111,042 raw | 8,156 indexed | 92.7% saved |
Wire it into your agent this afternoon: free tier is 2,000 reads a month, no card, and the key is live seconds after signup.
Get API access →First 1,000 builders: Starter free for 3 months