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
457,778 tools. Updated 2026-08-14 15:38
"LLVM" matching MCP tools:
- Classify text into user-supplied labels using a local Ollama model, validating replies against the label set and reporting matched labels or declined responses when no label fits.MIT
- Get project sequence details (name, prefix, pad, last_used, next, next claim ID) to avoid guessing conventions; ask before seeding missing sequences.AGPL 3.0
- List all participants in the project roster with their name, kind, lane, admin status, and token activity. Use this to check for duplicates before provisioning a new admin.AGPL 3.0
- Find available compiler versions by searching with a regular expression to match compiler names, returning detailed compiler information for development and analysis.MIT
- Transcribe local audio files (m4a, mp3, wav, ogg, flac) into text with faster-whisper. Get segmented and full transcripts, optionally translated.MIT
Matching MCP Servers
- AlicenseAqualityCmaintenanceAn MCP server that runs curated adversarial prompts against local Ollama models to test guardrails, scoring responses with heuristic verdicts for human review.5MIT
- FlicenseAqualityBmaintenanceMCP server to help AI assistants generate Italian PA user interfaces by providing components, recipes, accessibility rules, and validation for Bootstrap Italia and Web Components.9
Matching MCP Connectors
Forecast future events and scan prediction-market edges.
Swiss weather data for AI assistants — forecasts, measurements, stations, pollen.
- Create or append to markdown notes in a wiki vault without overwriting existing content. Use dry_run mode to preview changes before writing, while immutable source files stay protected.MIT
- Retrieve outbound wiki-links and backlinks for a note. Enter a file path to get all links pointing from the note, plus references from the rest of the vault matched by filename, title, or aliases.MIT
- Generate adversarial test tasks targeting LLM failure modes like hallucination, prompt injection, and reasoning errors, including expected behaviors for automated evaluation.MIT
- Scores an agent's answer against expected behavior for a prompt, returning pass/fail, a 1-5 score, and justification. Automates assessment of adversarial task responses.MIT
- Retrieve your resolved identity including actor, project, and admin status within the LLM Bus coordination layer.AGPL 3.0
- Analyze a per-line coverage report against the git diff to compute patch coverage and gate result, rendering the report as Markdown, YAML, or JSON.BSD 3-Clause
- Retrieve aggregated statistics on all recorded evaluations: total count, success count, pass rate, and average score. No API key required.MIT
- Revoke active tokens of a project participant by name, locking them out until re-granted, with history preserved.AGPL 3.0
- Analyzes images from URLs or base64 data, returning structured JSON with summary, text, and detected objects for one-time image analysis.MIT
- Engage in multimodal, multi-turn conversations that include images, with concise automatic replies. Ideal for interactive image discussions instead of structured analysis.MIT
- Perform in-depth web research on any query using ChatGPT Deep Research, returning comprehensive results with sources in 5-15 minutes. Note: blocks other ChatGPT operations during research.MIT
- List all Ollama models stored on disk with their sizes and identify which model the server uses by default, helping verify model availability before making requests.MIT
- Authenticate requests via JWT and bcrypt, then process JSON inputs through a LangChain-backed Gemini model on Vertex. Rejects unauthenticated calls.MIT
- Rotate a teammate's token by participant name. Mints a fresh token and revokes old ones in one transaction, preventing lockout gaps.AGPL 3.0