Auto-load skills and capabilities into agent sessions with MiniMax vision and search proxy integration. For developers building multi-skill AI agents that need dynamic capability injection and vision/search features.
io.github.divitkashyap/auto-skill-loader
Local install
STDIO
1 required env var
How models use it and what it is built for.
Auto-load skills and capabilities into agent sessions with MiniMax vision and search proxy integration. For developers building multi-skill AI agents that need dynamic capability injection and vision/search features.
Local install — runs as a subprocess.
Configuration this server reads at startup.
MiniMax Token Plan API key from https://platform.minimax.io
Where to find authoritative docs and source for Divitkashyap Auto Skill Loader.
Paste any of these into Agent Studio after connecting Divitkashyap Auto Skill Loader.
Common questions about connecting and running Divitkashyap Auto Skill Loader.
What does auto-skill-loader do?
It automatically loads and injects skills into agent sessions, providing MiniMax vision capabilities and search proxy functionality. This lets you dynamically extend agent capabilities without manual configuration.
How do I authenticate with MiniMax?
Set the MINIMAX_TOKEN_PLAN_KEY environment variable with your API key from https://platform.minimax.io. This is required to enable vision and search features.
How do I install auto-skill-loader?
Run `uvx auto-skill-loader@1.0.1` to install via uv. Ensure MINIMAX_TOKEN_PLAN_KEY is set in your environment before starting.
Can I use this without MiniMax?
The MINIMAX_TOKEN_PLAN_KEY is required, so you need a MiniMax account. The server is specifically designed to integrate MiniMax vision and search capabilities into agent sessions.
What transport protocol does this use?
Auto-skill-loader uses stdio transport, meaning it communicates via standard input/output streams. This makes it compatible with MCP clients that support stdio-based servers.
MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 30+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.
Open Agent Studio