|
JamJet — Prove every agent action.
JamJet controls AI-initiated production changes and emits portable receipts your security, compliance, and audit teams can verify. Open spec, Python and Type...
|
86 % |
|
JamJet — Prove every agent action.
JamJet controls AI-initiated production changes and emits portable receipts your security, compliance, and audit teams can verify. Open spec, Python and Type...
|
86 % |
|
Agentist — Open-Source Harness for Production AI Agents
The open-source harness for production AI agents. One command runs your agent in a real sandbox — watched, spend-capped, recorded — with human approvals, coa...
|
83 % |
|
Tracefold — fork() a running agent. Replay it deterministically.
Tracefold is a fork-native microVM runtime for agent execution and RL rollouts. Snapshot, fork, branch, and deterministically replay a running execution.
|
83 % |
|
Oparento — Infrastructure for persistent AI agents
The runtime layer for persistent AI agents: execution, permissions, memory, identity, scheduling, sandboxing, and multi-agent coordination.
|
82 % |
|
GrubX | Runtime Action Firewall for AI Agents
GrubX checks every AI agent action, including MCP tool calls, API requests, and SaaS operations, against your policy before it executes. Allow, block, or req...
|
82 % |
|
Agentika - The Control Plane for Production AI
|
82 % |
|
AgentX: The Action Firewall for AI Agents
Block the catastrophic deterministically, coach the recoverable. AgentX is a runtime firewall for AI agents. It blocks catastrophic tool calls (DROP TABLE, s...
|
82 % |
|
CellaFlow | The coordination and durability layer for AI agents
The coordination and durability layer for AI agents. Many agents, one action: fenced idempotency leases admit a single caller, commit what it produced, and h...
|
82 % |
|
Runtime Guard - Any agent. Your rules.
Runtime Guard is an open-source MCP security layer for AI agent policy enforcement, runtime security, and destructive command protection. Runs local with no...
|
82 % |
|
Fuse AI | Block unsafe agent actions before execution
Fuse AI is a local-first runtime and diagnostic layer that checks AI agent tool actions before they hit APIs, files, databases, email systems, payment system...
|
82 % |
|
Nxolaryn AI | Runtime Recovery for Production AI Agents
Stop runaway AI agents before they hit production tools. Nxolaryn provides runtime recovery policies and decision records for production AI-agent workflows.
|
82 % |
|
Retrace, The Execution Replay Engine for AI Agents — Retrace
Record, replay, fork & share AI agent executions. See every LLM call, tool call, and error your agent makes, then debug in seconds. Free for 1,000 traces/mo.
|
82 % |
|
Hydaelyn · Durable agent workflows for Go
Hydaelyn is an embeddable Go runtime that persists tasks, leases, approvals, and side-effect records so long-running agent workflows can continue after a pro...
|
82 % |
|
AgenticosCore — Legacy Exit com prova antes do corte
Dual-run em paralelo, Equivalence Contract, exceções com dono, freio e Evidence Pack. Sem cortar produção no escuro.
|
82 % |
|
RunMesh — Durable runtime for AI agents
RunMesh is the durable runtime for long-running AI agents, with secure sandboxes, persistent state, retries, model failover, budgets, and complete run visibi...
|
82 % |
|
JoyMux — AI Execution & Runtime Security
JoyMux gives AI agents persistent sessions, parallel execution, compact context, and recovery—with identity, anomaly detection, containment, and runtime cont...
|
82 % |
|
Verify AI agent actions before tools run | ActionRail
ActionRail by ToolJet is an open-source runtime enforcement layer for AI agents. Check proposed tool calls against policy and live systems of record before e...
|
82 % |
|
Recuriant: the agentic operating system
Recuriant runs AI agents that write their own workflows. Every action passes a policy gate before it runs, a workflow an agent starts stays inside that agent...
|
82 % |
|
The Runtime for AI Labor — Cerebral OS
Cerebral OS enables you to deploy AI-powered workflows that execute real work across your production systems. Governed, auditable, and production-safe.
|
81 % |
Detected social media networks