Open source alternatives to OpenAPPA
200 open source apps like OpenAPPA, closest match first.
Top picks
- A

Adrian
LaunchOpen-source runtime AI agent security tool - monitors and controls AI agents, catching malicious tool use, prompt injection, and policy drift in real time, before the agent acts.
Self-hosted
- D

DashClaw
LaunchDashClaw is an approval layer for unattended AI agents. It stops risky agent actions before they run, sends them to you to approve or deny from the dashboard or your phone, and records every decision in a signed ledger. Policies decide whic
- O

Opengeni
LaunchOpen, self-hostable agentic service for organizations — durable, replayable agent sessions, human approvals, governed credentials and memory, running in managed sandboxes or on your own hardware. Apache-2.0.
Self-hosted
- L
Lacp
LaunchLocal-first policy, evidence, memory, and recovery controls for Claude, Codex, Hermes, and other CLI coding agents.
- O

OpenTaint
LaunchThe open source taint analysis engine for the AI era. A formal dataflow analysis tool you can customize and self-host, built so AI agents drive your application security analysis without burning tokens on every scan. AI-ready open source al
Self-hosted
- A

Agenta
LaunchAgenta is an open-source workspace for building and running AI agents. You create agents through chat, improve them with feedback, schedule automations and approve sensitive actions with human-in-the-loop controls. Each agent can have a com
Self-hosted
More like OpenAPPA
OpenLIT: Zero code Agent Observability
Continuous feedback for testing, tracing & fixing AI agents
The OSS LLMOps Stack
Modular, open source LLMOps stack that separates concerns: LiteLLM unifies…
Claw Orchestrator
Run Claude Code, Codex, Antigravity, Grok Build and OpenCode behind one runtime
Openclaw Runbook
Unofficial OpenClaw runbook for running agents day to day without burning money
About OpenAPPA
OpenAPPA
An information-flow policy engine for LLM agents
OpenAPPA is an information-flow policy engine for LLM agents. It is a deterministic guardrail that tracks how data moves across tool calls, rather than matching patterns or asking a second model to judge each call. The site says the approach is described in a paper that was accepted at NeurIPS. It also says OpenAPPA is open, vendor-agnostic and MIT licensed, and that it works with coding agents such as Archestra, Claude Code, kAgent, Pi and OpenCode. Policies are set up with batteries, which are ready-made rule sets for operations, validation, observability and reporting, and you can create your own. The docs include a playground, a deep dive, policy configuration and evaluation benchmarks. The site reports its own benchmark results, including task completion and attack success rates compared with Claude's auto mode and Microsoft's FIDES. The code is on GitHub under archestra-ai/openappa.
- Price
- Open source
- Category
- AI
- Website
- openappa.com


