Dyno Lab — an AI safety, alignment and interpretability research workbench for Apple Silicon. Explore activations, probes, SAEs and interventions with local inference, a Python SDK, APIs and MCP.
- ai
- ai-interpretability
- ai-safety
- aisafety
96,838 visible repositories in the index
457–480 of 644 · Most stars
Dyno Lab — an AI safety, alignment and interpretability research workbench for Apple Silicon. Explore activations, probes, SAEs and interventions with local inference, a Python SDK, APIs and MCP.
Let your AI agents call your phone over VoIP, and call them back. Your phone rings like a real call, even locked; a voice AI talks it through and returns the transcript. Self-hosted, open source, no phone numbers or Twilio.
AI Agent Identity and Access Management and Inference Firewall
TranslatedOne zero-dependency MCP gateway CLI for every server and agent skill. Token optimization, tool discovery and context compression: 71,929 -> 581 tokens (-99.2%, measured), schemas stay out of context. One config for Claude Code, Codex, Cursor, every agent. 561KB, pure Python. | Zero-dependency CLI: manages all MCP tools and skills, and tool discovery saves 99.2% of tokens.
Repository for the PyOpenGL Project
TranslatedAn intelligent money-saving assistant based on McDonald's MCP
TranslatedA Claude Code and Codex plugin that creates a draft of Korean-app-style mobile screens in about 15 minutes from a single-page PRD
Debugging manhole for python applications.
TranslatedA Bilibili command-line tool designed for AI Agents: QR code login, watching videos, commenting, direct messages, posting updates, uploading videos, all with JSON output
MicroPython drivers for Waveshare e-paper modules
A Full Featured GPS NMEA-0183 sentence parser for use with Micropython and the PyBoard embedded platform
CRAB (Clean Rebuild of Anything's Behavior): rebuild software from behavior, not source code. An MCP-powered toolkit for black-box observation, clean-room reimplementation, and behavioral verification.
AI security review CLI: SAST, secret scanning, SCA, threat modeling and architecture review with Claude Code, Copilot, Codex, Gemini, Cursor or an API key. Reads your design docs, Confluence and Jira through MCP. SARIF output for CI.
TranslatedMCP desktop companion for AI agent status, usage tracking, and everyday tools. Anything Universe · Desktop companion and entry point to everyday tools.
TranslatedCurated agent skills for social media: Xiaohongshu (RedNote), Douyin, WeChat, X, YouTube, TikTok and LinkedIn. Also covers self-media operations, including WeChat Official Accounts, WeChat Channels and Bilibili: topic selection, writing, images, publishing and data. Usable by agents such as 千问 (Qianwen, "Thousand Questions"), 豆包 (Doubao, "bun"), 龙虾 (Longxia, "lobster") OpenClaw, WorkBuddy, Claude Code and Codex.
dogpile.cache is a Python caching API which provides a generic interface to caching backends of any variety
Neuro-symbolic agents with verified memory. MCP math tools, an algebra repair Skill, and an open challenge. Verify solutions, learn checked rules, and reuse.
Autonomous OODA-loop pentesting agent. Hermes brain, Kali + Playwright tools over MCP, ATT&CK as the Orient language.
Efficient sample rate conversion in python
Open Lakehouse Compute Fabric: an open-source lakehouse platform with Microsoft Fabric Compute Extension.
🦙 Ollama interfaces for Neovim
TOMC Memory: CPU context compression and task memory for Codex, Claude Code, Cursor, TRAE, WorkBuddy and other MCP clients.
Automated Tactics Techniques & Procedures
The repository host gives no description.