Sentrino-mcp
Agentic AI pipeline that searches financial news, scores sentiment, and delivers an interactive price-volume-sentiment score dashboard — triggered by a single command inside Claude.
Gehosteter MCP-Server
npx add-mcp 'https://sentrino-mcp.onrender.com/mcp'Installiert in Claude Code, Codex, Cursor und mehr
Dokumentation
Sentrino — Financial News Sentiment Analytics
from Sentimeter-lab
Agentic AI pipeline that searches financial news, scores sentiment, and delivers an interactive price-sentiment dashboard — triggered by a single command inside Claude.
What it does
Sentrino fetches publicly available financial news for any (yahoo fianace) stock ticker (or FX ticker), scores sentiment of each article, and overlays the scores on a price-volume chart — producing a self-contained, downloadable HTML dashboard. No coding required.
One instruction in. One dashboard out
/Sentrino run
Ticker: <ticker.exchange>
Company nmae: <compay name>
From date: <yyyy-mm-dd>
To date: <yyyy-mm-dd>
Key Features
-
Global multi-source aggregation of News + Filings + Analyst Notes — searches across internet news articles, regulatory filings company law websites (e.g.SEC, FCA, RNS) and analyst recommendations
-
Sentiment Scoring — financial sentiment with signed score (Positive / Negative / Neutral) and probability breakdown
-
Price-Volume-Sentiment downloadable and shareable HTML Dashboard — interactive Chart.js HTML file with colour-coded sentiment sized dots on a price-volume chart, news table with clickable links, author and source, and article summary panel
-
Three-way Interactivity — chart, news table, and summary panel are fully linked; hover on any element highlights the others
-
News article Summaries — tone-preserving summaries for every article, displayed in the dashboard summary panel
-
Multi Model Porting — MCP can be used across multiple AI chat models though the output has only been tested for Claude
Note: All the news articles, research analyst reports and company law filings are collectively referred to as 'news items' or 'news articles' through this document for ease of reference
Who it is for
| Global user | Utility and value draw |
|---|---|
| Retail & HNI investors, Invetment Banks, Investment/ Portfolio Managers | Understand financial news sentiment context behind price and volume moves |
| Compliance analysts | Cross-reference sentiment change and price-volume change with news article publish timing |
| Financial researchers | Assess and trace impact of research reports on ticker price and volume across multiple venues |
Requirements
-
A Claude account (free tier supported — one custom connector included)
-
A personal API key and orchestrator md file (provided on trial and same continues on subscription, so no new set up on subscription)
-
No other software, API keys, or coding required
Pricing
| Plan | Price | Includes |
|---|---|---|
| Free Trial | 5 days | Full access, personal API key |
| Individual | £10 / month | Full access, personal API key |
Getting Trial Access and Subscription
Lemon Squeezy Link
- Use this link for trial followed by subscription. Subscriber needs to share the card details and subscription starts immediately after the trial unless specifically cancelled by the subscriber
5 day freetTrial and subscription link
Individual Access (without sharing card details)
-
Email sentrino@sentimeter-lab.com with subject line:
Sentrino Trial Request -
You will receive your API key, setup instructions and subsription link (payment is handled via Lemon Squeezy) within 24 hours The same API key will be activated for further use after subscription payment is received.
Enterprise Access and Licensing
- For enterprise deployment, white-labelling, custom integration into your organisation's AI environment (including non-Claude enterprise AI chat models), or licensing enquiries, email sentrino@sentimeter-lab.com with subject:
Sentrino Enterprise customisation request.
Setup (5 minutes)
1 — Add the MCP connector in Claude
- Go to claude.ai → profile icon → Settings → Connectors
- Click Add custom connector
- Paste the server URL:
https://sentrino-mcp.onrender.com/mcp - Click Add
2 — Create a Project
- Click Projects → New project → name it
<Company, Security or ticker name>
Tips: Opening a unique standalone Project for each of your watchlist Company or security helps you maintain a focussed persistent chat, carry out incremental research, ask predictive questions and saves you tokens
- Open the project → click Add instructions
3 — Paste your instructions
You will receive two files in your setup email: MCP_config.md and SKILL_00_orchestrator.md.
- Open
MCP_config.md— replace{BUYER_API_KEY}with your personal key (no quotes) - Copy the full contents of
MCP_config.mdand paste into the instructions box - Paste the full contents of
SKILL_00_orchestrator.mddirectly below it - Click Save
4 — Enable the connector
Inside your Sentrino project, start a New conversation → click + (bottom left) → Connectors → toggle Sentrino ON
5 — Run
/Sentrino run
Ticker: <ticker.exchange>
Company nmae: <compay name>
From date: <yyyy-mm-dd>
To date: <yyyy-mm-dd>
Output
A single self-contained .html file delivered in the chat — open in any browser
WIZZ.L 2026-03-11 to 2026-05-31.html
click the link for the interactive dashboard WIZZ.L 2026-03-11 to 2026-05-31.html
Note: in case the above link does not render properly then visit the github repo and dowload the file with the same name
Example prompts:
i) First prompt as above
ii) Extend the coverage period
Run the Sentrino for NVDA for the period from 2022-03-01 to 2022-05-31 AND and add it to the same dashboard generated above.
iii) Adding news links sourced from other AI chat models
Add the news links below to the dashboard
iv) Since the news, sentiment and price-volume data is in the same project instance, you can ask questions such as below.
list top 3 news articles which witnessed the most price impact before the publishing of the news article
v) Predict price range
Given the Sentiment built over last x months based on the news articles, company filings and research analyst recommendations, what is a price range one can expect in next 2 days. Keep your response grounded only based on the information in this chat instance.
Tips: i) Keep the date range to 4–6 weeks per run. Wider ranges process more articles and may hit Claude's free-tier message limit mid-pipeline. For longer periods, run in shorter batches.Ignore these instructions if you have claude subscription ii) Hosted on Render (always-on) - First response after inactivity may take up to 60 seconds — Claude will wait automatically
Troubleshooting
| Issue | Fix |
|---|---|
Error 401 | API key entered incorrectly — paste it again with no quotes or spaces |
| Server slow on first call | Wait 60 seconds — it wakes automatically |
| Pipeline stops mid-run | Date range is too wide — shorten to 4–6 weeks and retry or wait till your free-tier tokens are reinstated. Ask for completing the process terminated in the chat window |
| No articles found | Try another date range or different company ticker of the same company on another exchange for a test run. It is likely that there is no news on the company |
Changelog
| Version | Notes |
|---|---|
| 1.0 | Initial complete agentic Sentrino skill roll-out for subscription via Lemon Squeezy |
Roadmap - customization opportunities
i. Tick-by-tick price granularity
ii. Timezone-aware and news items timestamp granularity
iii. Comparison with OHL (open, high, low) and weighted average (wa) prices
iv. Sectoral or Industry news coverage
v. Non-English news coverage and sentiment scoring
vi. Comparison with derivative and index price-volume including sector and market indices and competitor tickers
vii. Develop correlation scores between sentiment and price-volume change
viii. Market Greek indicators overlay
ix. Social media and blog sentiment overlay
x. BYOI — Bring Your Own Intel (upload proprietary or subscribed research report or analysis)
xi. Overlay of the organisation trading (order and execution) data and integrating it with enterprise AI Chat model(s) and case management systems
Contact
General enquiries including custamization, enterprise adoption,Licensing and suggestions for improvement: sentrino@sentimeter-lab.com
You can also send in pull request in the github repo Github Repo
Licence
Copyright © 2026 Sentimeter-lab.com. All rights reserved.
This software, its associated skill instructions, MCP server, and all related intellectual property are proprietary and confidential. Access is granted solely to authorised subscribers under the terms of the Sentrino Subscription Agreement.
Permitted: Personal use by the licensed account holder. Prohibited: Redistribution, resale, reverse engineering, sharing of API keys, or reproduction of skill instructions in any form.
Unauthorised use constitutes infringement of intellectual property rights and may result in immediate licence termination and legal action.