{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_umbxwd39httv","handle":"envbert-mcp","url":"https://wellknown.network/agents/envbert-mcp","links":{"self":"https://wellknown.network/agents/envbert-mcp/record.json","html":"https://wellknown.network/agents/envbert-mcp","markdown":"https://wellknown.network/agents/envbert-mcp/record.md","api":"https://wellknown.network/api/v1/agents/envbert-mcp","status":"https://wellknown.network/api/v1/agents/envbert-mcp/status","claim":"https://wellknown.network/agents/envbert-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/envbert-mcp/claim.json","badge":"https://wellknown.network/agents/envbert-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/envbert-mcp/history","tools":"https://wellknown.network/api/v1/agents/envbert-mcp/tools"},"ard":{"identifier":"urn:air::server:envbert-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"envbert-mcp","summary":"MCP server for envbert EDD classification — stdio and streamable HTTP transports","description":"# envbert-mcp \n \nMCP server for **envbert** environmental due diligence (EDD) text classification — \nsupports both **stdio** (Claude Code, VS Code) and **streamable HTTP** (Claude.ai, \nany remote MCP client) transports. \n \nClassifies EDD text — site assessment reports, groundwater monitoring logs, \ngeological surveys — into categories like `Geology`, `Contaminants`, and \n`Remediation Standards` using a local DistilBERT model \n([d4data/environmental-due-diligence-model](https://huggingface.co/d4data/environmental-due-diligence-model)). \nNo LLM fallback, sub-second inference, and your data never leaves your machine. \n \n--- \n \n## Features \n \n- **Two transports, one codebase** — the same tool logic is exposed over stdio \n  and streamable HTTP; the model loads once per process and is shared across \n  all connections. \n- **Three MCP tools:** \n  - `check_envbert_status` — check whether the model is loaded and ready \n  - `classify_environmental_text` — classify a single passage \n  - `classify_environmental_document` — classify every paragraph of a document \n    concurrently, with a category breakdown and low-confidence flags \n- **Fast** — sub-second inference after warmup, no LLM round-trip \n- **Local-only** — the model runs on your machine; HTTP mode only binds to \n  `localhost` by default \n \n--- \n \n## Installation \n \n```bash \npip install envbert envbert-mcp \n``` \n \nVerify: \n \n```bash \npython -c \"from EnvBert.due_diligence import envbert_predict; print('OK')\" \nenvbert-mcp --help \n``` \n \n> **Note:** the importable module is `EnvBert` (capital E and B), not \n> `envbert`. \n \n--- \n \n## Usage \n \n### stdio (Claude Code / VS Code) \n \nDefault mode — no flags needed. Claude Code and the Claude Code VS Code \nextension spawn this as a subprocess automatically once configured \n(see below). \n \n```bash \nenvbert-mcp \n``` \n \n### Streamable HTTP (Claude.ai / remote clients) \n \n```bash \nenvbert-mcp --http \n``` \n \nRuns on `http://127.0.0.1:8000` by default. First call after startup takes \n1…","publisher":{"name":"Afreen Aman","url":null},"homepage":"https://github.com/AfreenAman/envbert-mcp","repository":"https://github.com/AfreenAman/envbert-mcp","version":"4.1.0","license":null,"protocols":["mcp"],"tags":["envbert","mcp","edd","environmental","due-diligence","claude","bert"],"pricing":null,"endpoints":[{"url":"pypi:envbert-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"ai.model-access","name":"Model Access","confidence":0.905,"provenance":"derived"},{"slug":"dev.monitoring","name":"Monitoring & Observability","confidence":0.802,"provenance":"derived"}],"categories":["ai","dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"envbert-mcp","registry":"pypi","observedAt":"2026-11-03T02:25:10.782Z","publishedAt":"2026-07-15T14:19:10.551914Z","latestVersion":"4.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"envbert-mcp","url":"https://pypi.org/project/envbert-mcp/","firstSeenAt":"2026-09-09T15:21:36.316Z","fetchedAt":"2026-11-03T02:23:41.729Z","normalizedAt":"2026-11-03T02:23:41.729Z"}]},"firstSeenAt":"2026-09-09T15:21:36.316Z","updatedAt":"2026-11-03T02:25:10.782Z"}