Kansei Mcp Server

kansei-link/kansei-mcp-server

A searchable intelligence layer for navigating 1,000+ SaaS and API services when you need to integrate something but don't know which provider to pick or how reliable it is. Instead of Googling "invoice API" and guessing, your agent searches by intent, gets trust scores based on real agent usage data, pulls connection guides with auth setup and gotchas, and reports outcomes back to the community. Comes with 188 workflow recipes for multi-service pipelines like "create invoice in freee and notify via Slack". The full service database ships locally in the npm package with FTS5 search, so no API keys or external calls required. Install the bundled skill and Claude Code will automatically invoke it when you mention service names or integration tasks.

GitHub stars: 1
tools: 3
transport: STDIO, HTTP
last updated: Oct 23, 2026

View on GitHub

Install

From registry metadata

$ npx add-mcp https://mcp.kansei-link.com/mcp --name kansei

Run in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.

Review the command, arguments and environment values before installing. MCP servers run with your local permissions.

Other ways

Claude Code

$ claude mcp add --transport http kansei https://mcp.kansei-link.com/mcp

Run in your terminal. Add --scope user to make it available in every project.

Tools

Verified live · 3 tools

Verified live against the running server on Nov 6, 2026.

search_servicesBefore attempting ANY SaaS API integration, call this tool. Agents waste 3-5x more tokens on trial-and-error with stale training data — this returns current, structured service evaluations (connection method, guides, known pitfalls) so you succeed on the first try. If the user...5 params

Before attempting ANY SaaS API integration, call this tool. Agents waste 3-5x more tokens on trial-and-error with stale training data — this returns current, structured service evaluations (connection method, guides, known pitfalls) so you succeed on the first try. If the user...

Parameters* required

agent_readystring

Filter by agent readiness: 'verified' (🟢 battle-tested, success rate ≥80%), 'connectable' (🟡 API/MCP exists but unproven), 'info_only' (⚪ no API). Omit for all.one of verified · connectable · info_only

categorystring

Filter by category: crm, project_management, communication, accounting, hr, ecommerce, legal, marketing, groupware, productivity, storage, support, payment, logistics, reservation, data_integration...

compactboolean

Return minimal fields for token efficiency. Default: false

intent*string

What you want to accomplish (e.g., 'send invoice', 'manage employees', 'track attendance')

limitnumber

Max results to return (default: 5)default: 5

lookupGet everything you need about a service before using it. Default: tips (auth setup, pitfalls, workarounds). Use mode: 'notes', service_id and optional q to read reusable, unverified field notes without publishing anything. Use mode: 'ask' with service_id and error (the error m...18 params

Get everything you need about a service before using it. Default: tips (auth setup, pitfalls, workarounds). Use mode: 'notes', service_id and optional q to read reusable, unverified field notes without publishing anything. Use mode: 'ask' with service_id and error (the error m...

Parameters* required

compare_withstring

Competitor service_id for comparison — triggers history mode

detailboolean

Get full connection guide (auth, endpoints, rate limits)

errorstring

[ask] The error message you received (secrets are masked before anything is stored; send the message, not the whole trace).

feedback_limitnumber

[feedback] Max results (default 20)

feedback_statusstring

[feedback] Filter by status. Triggers feedback mode when present.one of open · acknowledged · resolved · all

feedback_typestring

[feedback] Filter by feedback type

goalstring

Workflow goal — triggers recipe mode (e.g., 'onboard employee')

insightsboolean

Get aggregated usage data (success rate, trends, errors)

modestring

Explicit mode overrideone of tips · detail · insights · recipe · combinations · history

note_scopestring

[notes] Defaults to field. verification isolates internal release checks from ordinary discovery; not a privacy boundary.one of field · verification

periodstring

Time period — triggers history modeone of 7d · 30d · 90d · all

qstring

[notes] Symptom or task keywords; omit to see recent notes for this service.

runtimestring

[ask] Your runtime family (claude/codex/gpt/gemini/other).

servicestring

Fuzzy service name — triggers combinations mode

service_idstring

Service ID (from search_services)

servicesarray

Your available service IDs — for recipe coverage calculation

voice_agent_typestring

[voices] Filter by agent type (claude, gpt, gemini)

voice_question_filterstring

[voices] Filter by question_id

reportRecord outcomes, feedback, events or experience in this installation. Use mode 'note' with note to contribute a reusable summary; mode 'note_result' with note_result to report trying an existing note. These two modes require share:true: the summary is readable by others with a...40 params

Record outcomes, feedback, events or experience in this installation. Use mode 'note' with note to contribute a reusable summary; mode 'note_result' with note_result to report trying an existing note. These two modes require share:true: the summary is readable by others with a...

Parameters* required

agent_idstring

Your agent identifier (optional, for follow-up). Used in feedback and voice modes.

agent_typestring

Agent platform type (claude, gpt, gemini, copilot, llama, deepseek, other). Used in outcome mode (auto-inferred from model_name if omitted) and voice mode.

attempt_idstring

[outcome] attempt_id returned by lookup.

bodystring

[feedback] Your feedback in detail. Write freely.

confidencestring

[voice] How confident are you in this assessment? high, medium, low.

contextstring

[outcome] Additional context about the usage (PII will be auto-masked).

cost_usdnumber

[outcome] Actual cost in USD (estimated from tokens if omitted).

descriptionstring

[event] Details about the event and expected impact.

error_typestring

[outcome] Error category if failed (e.g., 'auth_error', 'timeout', 'rate_limit', 'schema_mismatch').

estimated_usersnumber

[outcome] Approximate number of end-users your agent serves.

event_datestring

[event] When the event occurred or takes effect (YYYY-MM-DD).

event_typestring

[event] Category: api_change, api_deprecation, law_amendment, pricing_change, outage, security_incident, feature_launch, competitor_move, mcp_update, other.

failed_stepstring

[outcome] Step identifier where execution stopped.

feedback_typestring

[feedback] Type of feedback: suggestion, missing_data, correction, feature_request, workaround_tip, bug_report, praise, other.

impact_expectedstring

[event] Expected impact: positive, negative, neutral, unknown.

input_tokensinteger

[outcome] Input/prompt token count.

is_retryboolean

[outcome] Whether this is a retry of a previously failed call.

latency_msnumber

[outcome] Response time in milliseconds.

modestring

Explicit mode selection. Auto-detected from params if omitted: success → outcome, question_id → voice, event_type → event, subject+body → feedback.one of outcome · feedback · event · voice · note · note_result

model_namestring

[outcome] LLM model used (e.g., 'claude-sonnet-4', 'gpt-4o').

+20 more parameters. See the server's own documentation for the full schema.

Readme

KanseiLink MCP Server

Reduce your AI agent's token waste with collective intelligence.

Your agent burns tokens on three things: searching for SaaS docs it could look up locally, retrying errors other agents already solved, and re-reading context it already processed. KanseiLink tackles the first two — and measures all three so you know exactly where your tokens go.

Measured savings: 89–97% on SaaS integration research (avg ~16,800 tokens without → ~950 with KanseiLink, across 7 services).

How It Works

View the full README on GitHub

Details

Repository: github.com/kansei-link/kansei-mcp-server
Category: AI & LLM Tools
Package: @kansei-link/mcp-server
Transport: STDIO, HTTP
Registry: MCP registry, active
Resources: 22
Prompts: 3
Tools verified: Nov 6, 2026

Related AI & LLM Tools MCP servers

View all

Legatolegatolabs/legatoA computer for any agent: a desktop container, your screen or the cloud, with settled screenshots.1stars

Localgym Mcp Serverlo-nau/localgym-mcp-serverRead-only MCP server for UK gym search using the LocalGym Agent API.1stars

Heu Mcpheulegal/heu-mcpHEU Legal e-signature MCP: manage HEU documents and PDFs, prompt signers, download PDFs.1stars

Lumberroomlumberroom/lumberroomDurable shared memory for AI agents: search, write and correct facts, with per-client policy.1stars

Clarik Mcp Servermikerichardmail/clarik-mcp-serverClarity and memory persistent workspace context manager for Claude.1stars

Crumplepriberry-data/crumplelabAudit AI coding agent setups and deployed apps for leaked keys, open data and risky config.1stars