Search game code by what it does using local Ollama embeddings: index scripts, then query plain-language behavior to find matching chunks with paths and line numbers.
MIT
743,447 tools. Updated 2026-11-04 20:36
"A server for local file operations on Windows that integrates with Ollama" matching MCP tools:
- Starts a background watcher that automatically detects and dismisses CST dialog windows, preventing them from blocking long-running operations. Logs each dismissed dialog for later retrieval.MIT
- Search a built-in curated catalog of known local models by name, family, or task to find Ollama tags offline; use live model search when per-quant fit is needed.Creative Commons Attribution Non Commercial No Derivatives 4.0 International
- Inspect a code file using a local LLM to get written findings on security, performance, style, or structure, without modifying the file.MIT
- List free-seat status and requestable local runtimes such as Ollama, llama.cpp, LM Studio, OpenAI-compatible servers, and vLLM.MIT
- Upload a local file to attach to a submission comment. Returns a file ID; the file stays hidden until you call post_submission_comment with that ID in file_ids.MIT
Matching MCP Servers
- AlicenseAqualityBmaintenanceProvides a local MCP server for AI-powered web crawling on Windows, enabling agents to directly crawl and extract web content without requiring WSL.21-
- AlicenseBqualityAmaintenanceEnables comprehensive Windows system management through Claude Desktop, including PowerShell/CMD execution, file operations, archive handling, Git tools, network testing, and system monitoring. Provides a complete toolkit for Windows automation and administration tasks.596MIT
Matching MCP Connectors
On npm, Ollama's client fell from 27 to 6.6 downloads per 1,000 cloud lab SDK downloads in 46.
- Local MCPOAuth
Mac & Windows: let ChatGPT, Claude & Cursor use your email, calendar, iMessage, Teams, files.
- Upload a local file and get a file ID to attach to a submission comment. Nothing is shared until you post the comment with that ID.MIT
- List local Ollama models and retrieve queryable paid route labels for accessing AI models with payment options.MIT
- Send messages to a local Ollama model acting as a sovereign digital twin. Responses are WORM-sealed for immutable record keeping.Apache 2.0
- Check local GPU and Ollama availability, route coding tasks across three compute tiers, and track cumulative cloud API dollar savings based on token usage.MIT
- Run a prompt with an Ollama model by providing the model name and prompt text, with an optional system prompt for context.MIT
- Search notes by semantic meaning using local Ollama embeddings to find related content, even when no keywords match. Requires Ollama with nomic-embed-text; otherwise use keyword search.MIT
- List all installed Ollama models on your local machine and check their memory load status to discover available model names for chat, generation, or model info queries.MIT
- Download a remote file from a named server to a local path using SFTP. The operation is read-only on the remote side and overwrites any existing local file.MIT
- Send a prompt to a local Ollama model for immediate private answers. Use it for summarising text, drafting boilerplate, or extracting fields without sending data off-machine.MIT
- Search local Markdown docs for LangChain, LlamaIndex, Ollama, and XRPL concepts, symbols, and questions. Returns compact chunks without live crawling; shorten queries that yield zero matches.MIT
- Download a single file from a remote server to a local path via SFTP, overwriting local content, with read-only remote access.MIT
- Start a background watcher that automatically detects and dismisses CST dialog windows during long-running operations, preventing dialogs from blocking optimization loops or script execution.AGPL 3.0
- Upload a local file to a remote server over SFTP, replacing the destination file. Requires an existing local file and a writable remote path; for directory transfers use ssh_sync.MIT
- Install Ollama directly on macOS without Homebrew or sudo, enabling local inference for massive context processing. Supports downloading, starting the service, and pulling models like gemma3:12b for headless setups.MIT
- Run local AI inference using prism-coder models, automatically selecting the largest viable tier (2B-27B) based on available RAM. Falls back to cloud when local fails, saving tokens on successful runs.Apache 2.0