About Reecopedia
Reecopedia is an EU regulatory search service for textile sustainability and compliance work. It retrieves cited passages from legislation and official studies; the AI client composes answers, with no LLM running on the server.
Key features
- 01Search EU textile regulation corpus
- 02Return passages with document and page citations
- 03Browse document outlines
- 04Filter corpus documents by title
- 05Find exact text within a document
Use cases
- 01Find ESPR provisions and their sources
- 02Research textile Digital Product Passport requirements
- 03Locate an article number or definition
MCP support
The remote MCP server exposes regulatory search, document-reading, and text-location tools. Retrieved passages include citations, while the client generates the answer.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://ia.reeco.eco/mcpclaude mcp add --transport http reecopedia https://ia.reeco.eco/mcp{"mcpServers":{"reecopedia":{"type":"http","url":"https://ia.reeco.eco/mcp"}}}Open in Cursor{"servers":{"reecopedia":{"type":"http","url":"https://ia.reeco.eco/mcp"}}}Open in VS Code{"mcpServers":{"reecopedia":{"type":"http","url":"https://ia.reeco.eco/mcp"}}}{"mcpServers":{"reecopedia":{"serverUrl":"https://ia.reeco.eco/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
8| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | Yes | |
| scope | any | No | |
| top_k | integer | No |
Details
From the project READMEReecopedia MCP Server
Free regulatory search engine for EU textile sustainability law, exposed as a remote MCP server. No API key, no account, no installation — connect any MCP client to:
https://ia.reeco.eco/mcp
Published in the official MCP Registry as eco.reeco/reecopedia (domain-verified). Interactive web version: ia.reeco.eco/reecopedia.
What it covers
Reecopedia answers questions from a curated corpus of EU textile sustainability law and official preparatory studies:
- ESPR — Ecodesign for Sustainable Products Regulation (EU) 2024/1781
- Digital Product Passport (DPP) for textiles
- CSRD — Corporate Sustainability Reporting Directive
- CBAM — Carbon Border Adjustment Mechanism
- EU ETS — Emissions Trading System
- CWA 18291 — CEN Workshop Agreement (CIRPASS-2)
- JRC preparatory studies for textile ecodesign
Every answer is grounded in retrieved passages with source citations (document, page). Retrieval only — your AI client writes the answer; no LLM runs on Reecopedia's side.
Tools
| Tool | Purpose |
|---|---|
search_eu_textile_regulations | Main search over the regulatory corpus, returns cited passages |
search / fetch | Same corpus in the deep-research format (search by query, fetch full document page by id) |
document_outline | Table of contents of one document, to navigate a long act or standard |
list_corpus_documents | List available corpus documents, optionally filtered by title |
find_exact_text | Locate a literal string (e.g. an article number or defined term) inside one document |
report_answer_issue | Flag a wrong or unsupported answer for correction |
about_reecopedia | What Reecopedia is, who runs it, privacy policy |
Connect
Claude (claude.ai / Desktop) — Settings → Connectors → Add custom connector → URL https://ia.reeco.eco/mcp
Claude Code
claude mcp add --transport http reecopedia https://ia.reeco.eco/mcp
ChatGPT (developer mode) — Settings → Connectors → Add custom connector → URL https://ia.reeco.eco/mcp, no authentication.
Cursor — add to .cursor/mcp.json in the project, or ~/.cursor/mcp.json for every workspace:
{
"mcpServers": {
"reecopedia": {
"url": "https://ia.reeco.eco/mcp"
}
}
}
VS Code (GitHub Copilot) — add to .vscode/mcp.json in the workspace, or run MCP: Open User Configuration:
{
"servers": {
"reecopedia": {
"type": "http",
"url": "https://ia.reeco.eco/mcp"
}
}
}
Gemini CLI — install as an extension, one command:
gemini extensions install https://github.com/halleluiaman-ux/reecopedia-mcp
Restart Gemini CLI after installation, then run /extensions list and /mcp
to check that reecopedia is loaded and connected. Try:
Use Reecopedia to find Article 9 of Regulation (EU) 2024/1781. Fetch the relevant passage and cite its document title, reference and original URL.
The extension exposes seven source-reading tools. report_answer_issue is excluded
because it submits a report. Gemini CLI authentication is separate from the
Reecopedia server, which requires no account or key.
Or add it by hand to ~/.gemini/settings.json (httpUrl selects the streamable HTTP transport):
{
"mcpServers": {
"reecopedia": {
"httpUrl": "https://ia.reeco.eco/mcp"
}
}
}
Any other MCP client — add a remote MCP server with transport streamable-http and URL https://ia.reeco.eco/mcp. Generic JSON config:
{
"mcpServers": {
"reecopedia": {
"type": "streamable-http",
"url": "https://ia.reeco.eco/mcp"
}
}
}
Gemini API example
examples/gemini_api.py connects Google’s Interactions API to the same remote MCP endpoint using the official Python SDK. It requires your own Gemini API key and a model available to your Google account. Google’s API usage terms and charges apply separately from public Reecopedia access.
python3 -m venv .venv
.venv/bin/pip install -r examples/requirements.txt
# Set GEMINI_API_KEY securely in your environment.
.venv/bin/python examples/gemini_api.py --model YOUR_AVAILABLE_GEMINI_MODEL
The example uses a 300-second Google request timeout (configurable with
--timeout-seconds), low thinking, bounded output, no automatic retries, and restricts tools to source
reading, and exits unsuccessfully unless the response contains a matched MCP call
and result plus an answer. A plausible answer alone does not prove MCP use.
--check validates the request locally without a key or network call; it does
not verify Google access or a live integration.
API failures print a diagnostic JSON to stderr: HTTP status, structured API error status/message when provided, and an explicit allowlist of response trace headers. Request headers, URLs, cookies and arbitrary response bodies are excluded; the active API key is redacted from the selected values. A client timeout can have no HTTP status or response headers. Missing tracing headers remain absent.
Gemini CLI gallery discovery uses the GitHub topic gemini-cli-extension and
the root manifest. A valid repository does not guarantee gallery inclusion or
availability in Gemini web/mobile. See Google’s
extension release guide and
remote MCP API documentation.
Verification on 23 October 2026
Gemini CLI 0.65.0 validated and loaded extension 1.0.1 and reported the public
MCP endpoint as Connected. A separate MCP client completed a public search and
fetched the returned Article 9(2) identifier successfully. The API example passed
thirteen local contract tests using google-genai 2.33.0 and a mocked HTTP transport.
A live plain streaming Google request returned OK after 161.4 seconds.
The remote MCP stream began a reecopedia:find_exact_text call and result, but
ended without a completed answer; non-streaming MCP requests timed out even at
300 seconds. A completed Gemini API + MCP answer remains unverified.
These checks do not certify answer quality or gallery inclusion.
Privacy
The server receives only your query text — no account, no login, no personal data. Queries are stored without any personal identifier (no IP, no user id, no session) to improve retrieval quality, retained at most 90 days, then deleted. Full policy: ia.reeco.eco/reecopedia/privacy.
Who runs it
Built by Stefano Cipriani, founder of Reeco® — algorithmic verification layer for textile supply-chain compliance (Digital Product Passport, mass-balance reconciliation). Expert Member of CIRPASS-2 (EWG1, EWG3), JRC Registered Stakeholder.
Cursor plugin
This repository is also a Cursor marketplace repository. The plugin under
plugins/reecopedia adds the same remote server to Cursor;
the marketplace manifest is .cursor-plugin/marketplace.json.
About this repository
This repository documents the public Reecopedia MCP endpoint (server.json mirrors the official MCP Registry entry). The retrieval engine and corpus are not open source.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.