# Klavis-AI/klavis

Klavis AI:  MCP integration platforms that let AI agents use tools reliably at any scale

Repository: https://github.com/Klavis-AI/klavis
Canonical: https://ross.abutalabs.com/products/klavis
Homepage: https://www.klavis.ai/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: ai, discord, llm, mcp, mcp-client, mcp-server, open-source, agents, ai-agents, api, developer-tools, function-calling, integration, oauth2
Last push: 2026-06-01T21:21:04+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 85, release rhythm 56, longevity 36
- inputs: {"age_days": 506, "days_push": 93, "days_rel": 216, "gap_med": 0.0, "n_releases_24m": 79}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5793, forks 559 (observed 2026-08-28T04:09:29.678196+00:00)

## What it is
Klavis AI is an open-source MCP (Model Context Protocol) integration platform that lets AI agents connect to 100+ prebuilt tool integrations with OAuth support, plus Strata, a single MCP server that lets agents handle thousands of tools while optimizing context window usage. It offers hosted MCP servers, Slack/Discord clients, a web UI, and a RESTful API for integrating MCP functionality into applications.

## Use cases
- connect ai agents to external saas tools via mcp
- add oauth-secured tool integrations to an llm agent
- let one agent use thousands of tools without blowing the context window
- host mcp servers for my ai application
- add mcp tools to a discord or slack bot
- manage mcp integrations through a web ui
- expose tools to agents through a rest api

## When to choose
- you need many prebuilt MCP integrations with OAuth handled for you
- your agent struggles with too many tools and you want context-optimized routing via Strata
- you want hosted MCP servers instead of self-managing each integration
- you want to add tool use to Slack or Discord bots

## When to avoid
- you only need one or two simple tool integrations you can implement directly
- you require fully self-hosted everything and cannot use a hosted API
- you need a general agent framework rather than an MCP tool-integration layer

## Facets
- artifact type: service
- maturity: active
- function: mcp, agent-framework, auth, api-framework, sdk, developer-tools
- domain: large-language-models, apis, developer-tools
- platform: python, self-hosted, cloud
- tags: mcp-server, mcp-client, oauth2, function-calling, tool-use, hosted-mcp, integrations, strata, yc-startup, ai-agents, automation, web-server, docker, discord, slack

## Member repositories
- Klavis-AI/klavis (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:29.678196+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:52:48.039579+00:00, confidence not recorded.
  - readme: https://github.com/Klavis-AI/klavis (fetched 2026-08-28T04:09:29.678196+00:00, sha 72dd0a7e4685)
  - homepage: https://www.klavis.ai/ (fetched 2026-08-29T08:48:07.768106+00:00, sha 5e2378ae5046)
  - site_page: https://www.klavis.ai/docs (fetched 2026-08-29T08:48:07.776965+00:00, sha 5e1acb41a784)
  - site_page: https://www.klavis.ai/pricing (fetched 2026-08-29T08:48:07.779221+00:00, sha 04d0313248f0)
  - site_page: https://www.klavis.ai/faq (fetched 2026-08-29T08:48:07.781436+00:00, sha 420a7731e496)
- Data as of 2026-08-30T08:39:29.467469+00:00.