swival MCP Server
Swival/swival
[](https://agentrank-ai.com/tool/Swival--swival/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) A small, powerful, open-source CLI coding agent that works with open models.
claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
Swival/swival is a Python tool licensed under MIT. A small, powerful, open-source CLI coding agent that works with open models. Topics: agent, ai, cli, code, eval, huggingface, lmstudio, coding, a2a, agent-to-agent, mcp.
Ranked #1128 out of 24846 indexed tools.
In the top 5% of all indexed tools.
Ecosystem
Score Breakdown
343 stars → growing community
Last commit 1mo ago → recently updated
30/32 issues closed → responsive maintainer
4 contributors → small team
No dependents → no downstream usage
319 weekly installs → early adoption
Weights: Freshness 20% · Issue Health 20% · Dependents 22% · Stars 10% · Contributors 8% · How we score →
How to Improve
Badge all embed codes →
[](https://agentrank-ai.com/tool/Swival--swival/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) <a href="https://agentrank-ai.com/tool/Swival--swival/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge"><img src="https://agentrank-ai.com/api/v1/badge/Swival--swival" alt="AgentRank"></a> Matched Queries
From the README
# Swival A coding agent for any model. Swival makes it easy to work with local models and online services in the same tool. Smaller models get the same attention as the largest ones, with careful use of the space available for conversation and working notes. A short setup guide helps connect the model, and switching models doesn't mean changing agents. For online services, secret encryption can be enabled with a single option to hide supported API keys and tokens before they're sent to the model. Filters can also be added to remove private details, such as customer names and internal URLs. Longer jobs don't need constant reminders: Swival keeps notes on what it's learned and what's left to do. An interrupted session can pick up where it left off, and setting a goal keeps Swival working toward it. Checking the work is built in too. A single option adds a review step, so Swival can check its work and try again when the review finds problems. Another model can handle the review insteRead full README on GitHub →
Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.