agent-safe-pipeline
A blueprint for keeping AI agents from taking real-world actions on their own, by forcing every action through an independent policy check and a human approval step.
View decionis/agent-safe-pipeline on GitHub- GitHub stars
- 671
- Forks
- —
- Language
- TypeScript
Project details
- Repository
- decionis/agent-safe-pipeline
- Licence
- Apache-2.0
- Primary language
- TypeScript
- Catalog status
- Available
- Catalog check
- 2026-11-13
Repository description
Open-source gateway that puts authorization in front of every action your AI agents, workflows and pipelines take, before it commits: payments, trades, price changes, deployments and API calls run once on a single-use grant, wait for a person, or are refused, as Decionis decides, with chained evidence. Homebrew, Linux, Docker, Helm.
Explore by topic
More tools in AI agents
- agentes-chat
agentes-chat · AGPL-3.0
A self-hosted, open-source team chat app where AI agents (Claude, ChatGPT, Gemini, Grok) join your channels as real teammates, with their own computer and browser, and…
- langgraph-harness
vrajpal-jhala · MIT
A self-hosted AI agent platform for GitLab that reviews merge requests, resolves issues autonomously, and chats with full project context, already handling nearly 1,000…
- paperclip
paperclipai · MIT
An open-source dashboard where you hire a team of AI agents (Claude Code, Codex, Cursor, OpenClaw, etc.), give them a business goal, set budgets, and watch them work…
- librechat
librechat-ai · MIT
LibreChat is an open-source, self-hosted ChatGPT clone that connects to almost any AI model provider and adds agents, MCP tools, code execution and multi-user accounts…