Search Results for "safe"
Sort By:
Showing 63 open source projects for "safe"
View related business solutions-
Try Google Cloud Risk-Free With $300 in CreditUse your credit across every product. Compute, storage, AI, analytics. When it runs out, 20+ products stay free. You only pay when you choose to.
-
Custom VMs From 1 to 96 vCPUs With 99.95% UptimeLive migration and automatic failover keep workloads online through maintenance. One free e2-micro VM every month.
-
1
MCP Text Editor
Provides line-oriented text file editing capabilities
The MCP Text Editor Server provides line-oriented text file editing capabilities through a standardized API, optimized for integration with Large Language Models (LLMs). It enables efficient partial file access, minimizing token usage while ensuring safe concurrent editing. -
2
AgentScope Java
Agent-Oriented Programming for Building LLM Applications
...It provides a comprehensive toolkit for creating autonomous agents that can plan, execute, and adjust complex workflows, making decisions about which tools to invoke and how to solve multi-step problems. The framework includes runtime controls such as safe interruption and graceful cancellation to manage agent execution robustly in production environments. It also supports human-in-the-loop intervention, allowing developers or users to inject guidance at any point during reasoning while preserving state and tool context. Built with enterprise needs in mind, AgentScope Java integrates into traditional Java stacks and provides structured abstractions for memory, workflows, and tool invocation. -
3
OpenCode
The open source coding agent
OpenCode is an open-source AI coding agent designed to assist developers directly from the terminal, desktop, or IDE environments. It functions as a locally running assistant that can analyze codebases, execute development tasks, and interact with external tools while remaining highly configurable. The system is built around an agent model that can perform actions such as file manipulation, shell execution, and web access with user confirmation. Its provider-agnostic design allows developers... -
4
BrowserTools MCP
Monitor browser logs directly from Cursor
Browser Tools MCP is an MCP server and Chrome extension that gives AI agents safe, structured access to your live browser for debugging and automation. It can capture console/network logs, DOM snapshots, and screenshots, and expose them as typed resources the agent can query or act on. The design aims to make IDE agents (e.g., Cursor, Claude Desktop) more “web-aware,” enabling workflows like reproducing a bug, collecting evidence, and proposing fixes without copy-pasting. -
Catch Bugs Before Your Customers DoMove from alert to fix before users notice. AppSignal monitors errors, performance bottlenecks, host health, and uptime—all from one dashboard. Instant notifications on deployments, anomaly triggers for memory spikes or error surges, and seamless log management. Works out of the box with Rails, Django, Express, Phoenix, Next.js, and dozens more. Starts at $23/month with no hidden fees.
-
5
PKU Beaver
Constrained Value Alignment via Safe Reinforcement Learning
...These annotations include categories such as harmful language, unethical behavior, privacy violations, and other sensitive topics. By incorporating constraint-based optimization methods, Safe-RLHF trains models that balance reward objectives with safety requirements, ensuring that harmful outputs are penalized during training. -
6
OpenAI Codex CLI
Lightweight coding agent that runs in your terminal
...The CLI can scaffold new files, run code in sandboxed environments, install dependencies, and commit changes automatically, streamlining chat-driven development. It supports various approval modes—from suggestion-only to full automation—ensuring safe and controlled code execution. Codex CLI can also handle multimodal inputs like screenshots and diagrams to implement features intelligently. The tool includes built-in sandboxing & security measures, such as network restrictions and directory confinement, to protect your system during code execution. With extensive configuration options, including multiple AI providers and custom guidance files, it fits seamlessly into developer environments. -
7
shadowhook
Android inline hook library which supports thumb, arm32 and arm64
...The library allows you to specify hook targets either by function address or by library name + function name, and it automatically handles newly loaded shared libraries (ELFs), ensuring hooks remain effective even when code is dynamically loaded at runtime. ShadowHook strives to be stable and safe for production use: it automatically avoids circular-call recursion between original and proxy functions, and supports unwind/unwind-safe proxy functions (CFI/FP unwind) so that backtraces, debugging, and stack unwinding remain correct. It also offers optional intercept and logging capabilities, letting developers record hooked calls. -
8
TypeChat
Library for building type-safe natural language interfaces with LLMs
TypeChat is an open source library developed by Microsoft that simplifies the creation of natural language interfaces by using type definitions to structure interactions with large language models. Traditional natural language interfaces often relied on complex decision trees to interpret user intent and gather required inputs. With the rise of large language models, developers can interpret user requests more easily, but they still face challenges related to output reliability, safety, and... -
9
Anthropic SDK TypeScript
Access to Anthropic's safety-first language model APIs
anthropic-sdk-typescript is the TypeScript / JavaScript client library for the Anthropic REST API, enabling backend or Node.js usage of models like Claude. It wraps API endpoints for creating messages, streaming responses, and managing parameters in a type-safe TS environment. The library is designed for server-side use, interfacing with REST, and is stable for integration in web services or backend agents. Example usage shows how to instantiate the Anthropic client, call client.messages.create(...), and obtain responses. It supports streaming endpoints as well. Because TypeScript provides type safety, it helps avoid common errors in JSON interplay. ... -
Gemini 3 and 200+ AI Models on One PlatformBuild generative AI apps with Vertex AI. Switch between models without switching platforms.
-
10
Groq Python
The official Python Library for the Groq API
...Through this library, you can call Groq’s models from Python code — for example to request chat completions, code generation, transcription, or any supported endpoint — using idiomatic Python syntax. The SDK handles authentication (via environment variable or parameter), defines proper type-safe request/response data types, and supports both synchronous and asynchronous usage patterns depending on your application needs. This makes it easy to integrate Groq-powered AI capabilities into backend services, data pipelines, research notebooks, or applications written in Python. For those building AI-based tooling, automation scripts, or ML-backed backends, groq-python abstracts away HTTP request plumbing and exposes a clean API, accelerating development and reducing boilerplate. -
11
Burn
Burn is a new comprehensive dynamic Deep Learning Framework
Burn is a new comprehensive dynamic Deep Learning Framework built using Rust with extreme flexibility, compute efficiency and portability as its primary goals. Burn emphasizes performance, flexibility, and portability for both training and inference. Developed in Rust, it is designed to empower machine learning engineers and researchers across industry and academia. -
12
OpenAI Symphony
Symphony turns work into isolated, autonomous implementation runs
Symphony is an open-source framework designed to transform project tasks into autonomous implementation runs managed by AI coding agents. It allows teams to manage and prioritize work while the system automatically assigns coding agents to complete tasks. Instead of directly supervising AI agents, engineers can oversee higher-level workflows and project outcomes. Symphony integrates with project management tools to detect new tasks and initiate isolated environments where agents implement... -
13
agent-browser
Browser automation CLI for AI agents
agent-browser is a toolkit that embeds AI agent capabilities directly into the web browser, enabling agents to interact with web content, scripts, and user actions while maintaining security boundaries that respect user privacy and browser constraints. It effectively provides a sandbox where AI agents can read, scroll, click, and interpret pages in context, allowing them to automate workflows, answer questions about page content, or generate structured summaries directly from the user’s... -
14
NSFWJS
Client-side indecent content checking powered by TensorFlow.js
...It is powered by TensorFlow.js and the NSFW detection model, and delivers around 90% accuracy that is improving each time. NSFWJS classifies images with percentages under five categories, namely: drawing and neutral, which are both safe for work; sexy, which includes sexually explicit images; and hentai and porn, which are pornographic drawings and images. NSFWJS offers a 'browserified' version, an NSFW filter web extension that filters out NSFW images from your browser, and also has a separate React Native app. -
15
NanoClaw
A lightweight alternative to Clawdbot / OpenClaw
Nanoclaw is a lightweight, security-focused personal agent runtime designed as a slimmer alternative to larger “personal assistant” agent stacks, with an emphasis on being easy to audit and safe by default. It runs agent execution inside Apple containers to provide strong isolation boundaries, so individual chats and actions can be sandboxed with tighter filesystem and process separation than a typical single-process bot. The project connects directly to WhatsApp, letting you deploy an assistant that can chat in a familiar interface while still supporting real agent behaviors instead of simple call-and-response prompts. ... -
16
Matrix
Multi-Agent daTa geneRation Infra and eXperimentation framework
...That design makes Matrix particularly well-suited for large-batch inference, model benchmarking, data curation, augmentation, or generation — whether for language, code, dialogue, or multimodal tasks. It supports both open-source LLMs and proprietary models (via integration with model backends), and works with containerized or sandboxed environments for safe tool execution or external code runs. -
17
AutoMaker
Start directing AI agents
Automaker is an autonomous AI development studio designed to transform how software is built by allowing developers to describe features, then watching AI agents implement code, tests, commits, and more with minimal manual typing. Instead of writing every line of code by hand, users add feature cards to a Kanban board with natural language descriptions, and AI agents powered by the Claude Agent SDK handle multi-step tasks such as planning, generating code, running tests, and committing to an... -
18
Supabase MCP Server
Query MCP enables end-to-end management of Supabase via chat interface
An open-source MCP server that enables comprehensive management of Supabase projects through natural language interactions, providing capabilities such as SQL execution, schema management, and API integration. -
19
GitHub Agentic Workflows
GitHub Agentic Workflows
...By writing intent in markdown files, a developer can quickly generate .yml Actions workflows that perform tasks such as summarizing issues, automating triage, generating reports, or maintaining documentation, all without manually crafting YAML logic from scratch. The system emphasizes safety and guardrails, running agents in sandboxed environments with minimal permissions by default, and using “safe outputs” to constrain what the workflow can write back into the repository. It includes tooling for compiling, testing, and iterating on agentic workflows locally and integrates with GitHub’s existing Actions ecosystem. -
20
HexStrike AI MCP Agents
HexStrike AI MCP Agents is an advanced MCP server
HexStrike AI is an MCP server that lets LLM agents autonomously operate a large catalog of offensive-security tools. Its goal is to bridge “language models” and practical pentest workflows—enumeration, exploitation, vulnerability discovery, and bug bounty reconnaissance—under safe, auditable controls. The server exposes typed tools and guardrails so agent prompts translate to concrete, parameterized actions rather than brittle shell strings. It ships with curated tool adapters, task orchestration, and guidance for connecting popular agent clients (Claude, GPT, Copilot) to a hardened execution environment. Documentation highlights the breadth of supported utilities and positions HexStrike as a research and red-team aid, not a point-and-click exploit kit. ... -
21
ktrain
ktrain is a Python library that makes deep learning AI more accessible
...With only a few lines of code, ktrain allows you to easily and quickly. ktrain purposely pins to a lower version of transformers to include support for older versions of TensorFlow. If you need a newer version of transformers, it is usually safe for you to upgrade transformers, as long as you do it after installing ktrain. As of v0.30.x, TensorFlow installation is optional and only required if training neural networks. -
22
Open Interpreter
A natural language interface for computers
Open Interpreter is an open-source tool that provides a natural-language interface for interacting with your computer. It lets large language models (LLMs) run code locally (Python, JavaScript, shell, etc.), enabling you to ask your computer to do tasks like data analysis, file manipulation, browsing, etc. in human terms (“chat with your computer”), with safeguards. Runs locally or via configured remote LLM servers/inference backends, giving flexibility to use models you trust or have... -
23
GitHub MCP Server
GitHub's official MCP Server
The GitHub MCP Server exposes GitHub as a Model Context Protocol server so AI assistants can safely act on repos, issues, pull requests, gists, and more through a consistent tool interface. It’s designed to run locally or remotely and then be attached to MCP-capable clients (for example, Copilot Chat) so an LLM can search code, open files, create branches, draft PRs, label or triage issues, and query metadata without hard-coding GitHub APIs. The server defines tools and resources with... -
24
MemOS
AI memory OS for LLM and Agent systems
MemOS is an experimental operating system and runtime built around the concept of memory-centric computing, where memory objects are first-class citizens and program execution is organized around efficient, persistent memory access rather than traditional process and file system boundaries. The project explores rethinking system abstractions by tightly coupling computation with memory objects so that programs can operate on large datasets without expensive serialization or context switching.... -
25
Fractals
Fractals is a recursive task orchestrator for agent swarm
Fractals is an experimental open-source framework designed to orchestrate complex tasks using swarms of AI agents organized in a recursive structure. The system takes a high-level goal and decomposes it into a hierarchy of smaller subtasks, forming a self-similar tree that resembles a fractal structure. Each leaf node of this tree represents a specific executable task that can be processed independently by an AI agent. The framework runs these subtasks in isolated Git worktrees so agents can...
