Search Results for "approval"
Sort By:
Showing 32 open source projects for "approval"
View related business solutions-
$300 Free Credits to Build on Google CloudPut your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
-
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
1
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. ... -
2
HOL Guard
Open-source antivirus for AI agents
...It evaluates supported actions for secret exposure, prompt injection, unsafe commands, malicious packages, and risky MCP interactions. Safe operations can proceed automatically, while known threats are blocked and ambiguous actions can require human approval. The platform also examines plugins, skills, hooks, package installs, and agent configurations before they are trusted. It integrates with major coding agents including Claude Code, Codex, Cursor, Gemini CLI, OpenCode, and GitHub Copilot CLI. Local operation requires no cloud account, while optional cloud services add synchronized evidence, shared policies, fleet visibility, and team approvals. -
3
Munder Difflin
local multi-agent harness
...Each supported CLI runs as a real terminal process with its own identity, workspace, and visual avatar. A central orchestrator called Michael routes tasks, resolves routine coordination, and escalates sensitive actions for human approval. Agents communicate through local mailboxes, share a blackboard, and retain long-term memory across sessions. The app provides live terminals, task management, Git tools, cost telemetry, and optional per-agent worktrees. It supports major coding agents, bring-your-own keys, local LLMs, and desktop builds for macOS, Windows, and Linux. -
4
Deep Agents
The batteries-included agent harness
...Built on LangGraph, it includes persistence, checkpointing, streaming, and production-oriented orchestration. Agents can delegate to sub-agents, work with files, run shell commands, manage long contexts, and retain memory across sessions. It also supports human approval of tool calls, reusable skills, custom tools, and MCP servers. -
Build Data Resilience - Take the Assessment TodayIs your recovery strategy as strong as you think? Take this quick self-assessment to check your recovery readiness and gain tailored insights. In only 2 minutes, you'll learn where you fall on the recovery readiness scale.
-
5
Phantasm
Toolkits to create a human-in-the-loop approval layer
Phantasm offers toolkits to create a human-in-the-loop approval layer to monitor and guide AI agents' workflows in real-time, ensuring safety and reliability in AI operations. -
6
Cloudflare OS
Agent workspace built on Cloudflare Workers for creating documents
...Generated applications are called Gadgets and run as private, isolated instances that can be modified and securely shared. Gatekeepers control external access, manage authorization, log actions, restrict resources, and queue side effects for later human approval. Blueprints serve as reusable application templates rather than simple document templates. The system runs on Cloudflare Workers and is released as an early-access foundation that organizations can customize into their own company operating environment. -
7
Corsair
Your Agent's Integration Layer
...Corsair is designed for production agent systems that need OAuth, scoped authorization, human approval, and centralized integration control. -
8
Persona
Create agentic front-end experiences for the web in VanillaJS
...The widget is backend-agnostic and connects to any SSE-capable agent or model backend through the Persona streaming protocol. It supports streaming chat, multimodal attachments, voice input and output, WebMCP page tools, local client tools, approval gates, tool call visualization, artifact rendering, and safe markdown or HTML output. It also includes configurable layouts such as floating, docked, fullscreen, and inline experiences. Overall, it is useful for teams that want a customizable web assistant UI without committing to a specific framework. -
9
pool
pool is Poolside’s coding agent that runs in your terminal
...It can run as a standalone interactive terminal app, as an ACP server for compatible editors, as an ACP client connected to another agent server, or non-interactively through command execution. The tool supports slash commands, file and directory fuzzy search, shell mode, message rewind, and configurable approval modes. It integrates with open agent specifications such as AGENTS.md, Skills, MCP, and ACP, making it adaptable to different coding environments. Developers can use it for interactive coding, scripted tasks, CI-style checks, project analysis, and agent-assisted implementation. Its main value is providing a flexible coding-agent interface that can move between terminal use, editor integration, and automated execution. -
Go from Code to Production URL in SecondsSkip the Kubernetes configs. Cloud Run handles HTTPS, scaling, and infrastructure automatically. Two million requests free per month.
-
10
Agentspan
Durable, Distributed runtime for ALL of your agents
...It enables agents to persist beyond process crashes, allowing workflows to continue even after interruptions or restarts. The system is built for durability, meaning tasks can pause for extended periods, including waiting for human approval, and then resume seamlessly. It supports scaling across multiple environments, making it suitable for production-grade agent orchestration. Agentspan includes a local server and visual interface that allow developers to inspect execution flows and debug agent behavior. It also integrates with multiple model providers, enabling flexibility in selecting underlying AI systems. ... -
11
Agentic Inbox
A self-hosted email client with an AI agent, running entirely on Cloud
...The system supports full email functionality, including sending, receiving, threading, and attachment management through a modern web interface. Its AI agent can search conversations, summarize messages, and generate replies, while still requiring human approval before sending. The architecture combines durable state management, storage services, and AI inference into a unified workflow. Overall, it serves as a reference implementation for building AI-powered communication tools with full control over data and infrastructure. -
12
Hexis
Git-backed control plane for AI-agent skills, tools, context, identity
...A built-in remote MCP server lets compatible clients such as Claude, ChatGPT, Cursor, and Cline connect to the same governed workspace. Tool manuals can include encrypted secrets without exposing credentials directly to agents. Proposed skill changes can move through owner approval before they become available to the wider team. Hexis is self-hostable with Docker and is designed to keep agent capabilities portable across vendors. -
13
OpenClaw Studio
A clean web dashboard for OpenClaw
...It allows users to connect to an OpenClaw Gateway, monitor agents, and control workflows from a single location. The platform provides real-time chat capabilities, approval management, and job configuration tools for agent operations. Built with a control-plane architecture, it handles communication between the browser and the gateway through server-managed connections. OpenClaw Studio supports both local and cloud deployments, making it flexible for different development and production environments. ... -
14
Dao Code
Open-source TypeScript terminal coding agent for DeepSeek-V4
...The project emphasizes byte-stable prompts, prefix-cache reuse, and low-cost reflection or memory forks so longer coding sessions can remain affordable. It also includes cross-session memory that verifies saved facts against the current codebase instead of blindly trusting old context. Dao Code supports approval gates, slash commands, skills, MCP tools, hooks, custom subagents, permissions, crash recovery, shadow-git checkpoints, and autonomous long-task mode. It is aimed at developers who want a Claude Code-like workflow with stronger mainland China availability and DeepSeek-first economics. -
15
ASu-skills
Job-search workflow plugin suite
...It provides specialized entry points for building verifiable experience, improving resume content, creating resumes, preparing for interviews, and tracking applications. Its contributor workflow helps users identify realistic open-source contributions and prepares changes for approval before submission. Resume tools can rewrite experience for target roles and generate editable HTML layouts with PDF export. Interview workflows analyze resume claims, predict likely questions, and probe knowledge gaps through follow-up questions. The suite supports Claude Code, Codex, and TraeWork while sharing common skills, assets, and reference materials. -
16
Vendo
Embedded agents your customers use to automate work
Vendo is an open-source customization layer that embeds AI agents directly into B2B SaaS products. It lets customers build personalized views, micro-apps, and automations on top of a product without modifying its source code. The agent converts the host application's API into tools and executes them with the permissions of the signed-in user. Generated interfaces run inside sandboxed environments and can use the host product's visual components and branding. A centralized guard layer manages... -
17
AutoMaker
Start directing AI agents
...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 isolated git worktree. The core idea is to shift developers’ focus from mechanical coding to high-level architectural and product decisions while retaining control through review and approval of generated changes. Built with tools like React, Vite, Electron, and Express, Automaker offers both web and desktop workflows with real-time streaming of agent activity and visibility into progress. -
18
AOS Community Edition
AOS Community Edition: the open agent operating system
...Signed release metadata, checksums, provenance attestations, and runtime compatibility files strengthen supply-chain verification. Its MCP service gives Codex, Claude, and Grok a shared product edge with constrained approval handling. Forge tooling helps agents inspect capability gaps and build least-privilege capsules for new functions. -
19
TrueForge
The open-source agent harness, it turns LLM into a working agent
TrueForge is an open-source agent harness that provides the runtime layer needed to turn language models into working agents. It manages model calls, MCP tools, skills, sandboxing, approvals, context, and persistent session state. Agents can use OpenAI, Anthropic, Gemini, other catalog providers, or OpenAI-compatible endpoints. Human checkpoints support tool approvals, questions, and generative interfaces during execution. Context tools include subagents, deferred tool loading, compaction,... -
20
Magentic UI
A research prototype of a human-centered web agent
Magentic-UI is a research prototype developed by Microsoft that serves as a human-centered interface powered by a multi-agent system. It enables users to automate complex web tasks, such as browsing, form filling, and data analysis, while maintaining control over the process. The system emphasizes transparency and user involvement, making it suitable for tasks requiring both automation and human oversight. -
21
Claude Plugins Community
Community plugin marketplace for Claude Cowork and Claude Code
...It acts as the public mirror of Anthropic’s reviewed community plugin catalog. The marketplace metadata is synchronized nightly from Anthropic’s internal review pipeline. Listed plugins have passed automated security scanning and approval before distribution. Claude Code users can add the repository as a marketplace and install individual community plugins from it. Plugin submissions are handled through Anthropic’s submission process rather than direct repository pull requests. -
22
Moltis
A Rust-native claw you can trust
...The system supports multiple large language model providers alongside local models, enabling users to maintain privacy while still accessing cloud capabilities when needed. Moltis emphasizes security through sandboxed execution environments, where commands and browsing tasks run in isolated containers and require explicit approval. The platform also includes long-term memory powered by hybrid vector and full-text search, allowing the assistant to retain context across sessions. With multi-channel access such as web UI, Telegram, and API endpoints, Moltis functions as a unified automation hub intended for developers and advanced users who want full control. -
23
Wardrobe
Your clothes, extracted and organized with gpt-image
...It detects garments in imported images, creates clean product-style cutouts, and can generate modeled editorial previews using a reference photo. The web interface supports drag-and-drop, paste, editing, review, regeneration, and approval of generated results. Original images, processed assets, job data, and the JSON clothing library remain in a local data directory. Two bundled Codex skills automate clothing import and generate styled outfit collections from items already in the wardrobe. Users can run the workflow through Codex or the web interface. The project requires an OpenAI API key for AI processing and allows model and data locations to be configured. -
24
IP as Logo
Compact Agent Skill for simplified neo-skeuomorphic IP mascot logos
...The skill limits designs to a small number of basic shapes and normally uses three semantic colors. It proposes three design directions before generating six independent candidate images after approval. Composition rules favor large lower-corner crops, clear identifying features, and opaque square output. The skill follows the open Agent Skills format and can work across supported agents rather than depending on one specific AI product. -
25
Sol Advisor
Codex-native architect orchestration with Luna and Terra
...This separation is intended to reduce context contamination while preserving clear responsibility for quality and architectural decisions. The workflow can monitor implementation, inspect diffs, request corrections, coordinate dependent work, and control final approval. It is distributed as a Codex plugin with companion custom-agent definitions for the native execution path.
