AI & Machine Learning · Tisane + Claude
Tisane MCP with Claude: Automate Content Moderation and Feedback Analysis
The short answer
Claude works with Tisane through the Model Context Protocol, an open standard that acts as a universal adapter. Once connected with an API key, Claude can send text to Tisane, read the results for toxic content, topics and sentiment, and help teams decide what to do next, with approvals controlled by the organisation.
What is Tisane?
Tisane is a natural language processing service. It reads written text and detects problematic content, extracts the main topics people are discussing, and performs aspect-based sentiment analysis, which shows how people feel about specific parts of a product or service. It works across 27 languages. Trust and safety teams, community managers, customer experience leaders and product teams typically use it to handle large volumes of user comments, reviews and feedback.
How Claude connects to Tisane through MCP
The Model Context Protocol, or MCP, is an open standard that lets Claude securely use other software. In plain terms, it is a universal adapter between Claude and a tool. It lets Claude read information from the tool and take actions in it, using permissions that your organisation controls.
Once Tisane is connected through its MCP server or a secure managed connector, Claude can submit comments or messages for analysis, receive the flags for toxic content, pull out topics and read sentiment for each aspect mentioned. Claude can then summarise patterns, group similar feedback and suggest moderation actions in plain English.
Permissions and approvals keep people in charge. Access is limited to the minimum needed, and the sign-in uses the API key that your team controls. Consequential actions, such as removing content or escalating an account, can require a named person to approve them, and every step is recorded in audit logs.
What you can automate
Moderating user comments at scale
A community team receives thousands of comments each day. Claude can route new comments through Tisane, flag those with toxic or abusive language, and queue them for a moderator to review. Routine, clearly acceptable comments can pass through, which frees people to focus on borderline cases that need judgement.
Understanding product feedback by feature
A product manager wants to know what customers really think about each feature. Claude can analyse reviews and survey responses through Tisane, separating sentiment for battery life, price or design. The result is a weekly summary showing which aspects are improving and which are causing frustration.
Spotting emerging topics in support messages
Customer support leaders often miss new issues hidden in ticket volume. Claude can extract the main topics from incoming messages through Tisane and highlight any theme that is growing quickly, such as a fault that customers keep mentioning after a software update.
Monitoring multilingual communities
A business that serves customers in several countries needs consistent oversight in every language. Tisane covers 27 languages, so Claude can apply the same moderation and sentiment checks to feedback from different markets and report the results in one view for the central team.
Reporting on brand sentiment for leadership
Executives need a clear picture without reading every message. Claude can combine sentiment and topic results from Tisane into a short briefing that shows trends, notable shifts and examples, helping leaders make decisions based on what users are actually saying.
An end-to-end workflow: Moderated feedback to customer success
TriggerA new review or community comment is posted on the company's platform.
The message is sent to Claude, which passes the text to Tisane for analysis.
Tisane returns flags for toxic content, the topics mentioned and the sentiment for each aspect.
Clearly acceptable feedback is stored and tagged by topic in the company's CRM.
Messages flagged as abusive are held and a moderator is asked to approve or reject them.
Negative sentiment about a specific product aspect creates a ticket in the support system for the relevant team.
Claude prepares a weekly summary of topics and sentiment trends for the product lead.
OutcomeModerators spend their time on the cases that need them, customer teams see recurring problems quickly, and product leaders get a clear, evidence-based view of feedback. Results vary by organisation and volume.
How we set it up
Book a free automation audit
We review how your teams currently moderate content and handle feedback, and identify the workflows where Tisane and Claude would save the most time.
Connect Tisane with an API key
We set up the connection using the API key your organisation controls. Tisane can be reached through its MCP server or the direct API, so we choose the route that fits your systems.
Design the end-to-end workflow
We map how Tisane results flow into your other systems, such as your CRM, ticketing or Microsoft 365, and decide which actions need human approval.
Test against real past cases
We run historical comments and feedback through the workflow and check the results with your team, adjusting thresholds and rules until they match your standards.
Launch with monitoring and training
We go live with audit logs and performance monitoring, then train your moderators, analysts and managers to use the workflow confidently.
Who benefits
- Trust and safety lead
- Faster, more consistent flagging of harmful content, with human review kept for difficult cases.
- Community manager
- Less time spent on routine comments and more time engaging with members.
- Product manager
- Clear sentiment by feature, based on far more feedback than any team could read by hand.
- Customer experience director
- Early warning of recurring problems and a single view of feedback across languages.
Security and control
- Access uses an API key with the least permissions needed, and your team controls it.
- Consequential actions such as removing content or escalating accounts require named human approval.
- Every request and action is recorded in audit logs for review.
- Test results are checked against real cases before launch, and performance is monitored afterwards.
- Automated decisions are documented so that moderation choices can be explained to users and regulators.
For teams in the US, UK, India and the Middle East
Organisations in the United States and the United Kingdom have their own rules on data handling and online safety, and we design workflows to meet them. Indian organisations are subject to the Digital Personal Data Protection Act, 2023, and Middle East organisations, including those in the UAE and Saudi Arabia, must meet local data protection laws. We choose the hosting region for your custom application to meet your legal and contractual requirements, in India, the United States, the United Kingdom or the European Union.
Learn more about Claude and MCP
Anthropic's AI model and assistant.
Secure links between Claude and business tools.
How engineers build and change software with Claude.
Hosted agents that run multi-step work.
The open standard Claude uses to connect to tools.
Frequently asked questions
What does the Tisane MCP server do?
The Tisane MCP server lets Claude connect to Tisane's language analysis. Once connected, Claude can send text for analysis and receive results for toxic content, topics and aspect-based sentiment. The server acts as a secure bridge, so Claude can use these results in workflows your organisation controls.
Is connecting Tisane to Claude secure?
Security depends on the setup. We use the API key your organisation controls, grant only the permissions needed, and keep consequential actions behind human approval. Every request is recorded in audit logs, and we review the design with your IT team before launch.
What can Claude do with Tisane once it is connected?
Claude can submit comments or messages for analysis, read flags for problematic content, extract topics and review sentiment for specific aspects. It can then group feedback, summarise trends and suggest actions, while people approve any decision that has a real consequence.
How long does setup take?
Timing depends on how many systems are involved and how much testing is needed. A focused connection with one workflow is often quicker than a multi-system design. We confirm a realistic timeline during the free automation audit, once we understand your volumes and integrations.
Which API key or permissions are needed?
Tisane is accessed with an API key, which your organisation obtains and controls. We recommend a key with only the permissions the workflow requires, and we agree the exact access level with your team during the audit so that it matches your security policies.