|
AG-X | Embedding your AI agent with Self-Healing Mechanism
Give your AI agents an immune system. AG-X intercepts failures, generates autonomous vaccines, and ensures no mistake happens twice. One line of code. Produc...
|
84 % |
|
Sentrik — Authority Runtime for AI-Enabled and Autonomous Systems
A machine-enforceable control layer that determines whether an autonomous system or AI agent is authorized to take a proposed action — before the action exec...
|
84 % |
|
Kelhe — The runtime control point for AI agent actions
Most AI agents never leave the pilot because no one can trust what they'll do in production. Kelhe scores every action your agents take — before it runs — an...
|
82 % |
|
DriftGard — Agent Assurance for every release.
Test AI agents against reviewed business expectations, compare releases and catch regressions. Extend governance into execution with DriftGard Runtime.
|
82 % |
|
AI Agent Red Teaming & Prompt Injection Testing — okjul
Adversarial testing for AI agents. okjul red-teams your agent for prompt injection and tool misuse, then hands you reproducible evidence auditors accept.
|
82 % |
|
Me Runtime — the gatekeeper for your AI agents
Independent authorization layer for AI agents. The agent proposes, Me Runtime authorizes: green, yellow or red.
|
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 % |
|
AI Agent Security: Agent Execution Control Switch | Ackuity
Ackuity watches every AI agent action in real time and stops rogue moves before they run. It is the Agent Execution Control Switch for agentic AI security.
|
82 % |
|
Reflexes — runtime policy checks that make your AI agent behave
Reflexes is an open-source safety harness for AI agents. Write policies in plain English; it checks every action in real time — and when the agent crosses a...
|
82 % |
|
Agent Sentinel — Pre-Execution Safety for AI Agents
|
82 % |
|
ApprovalPlane — human approval for risky agent actions
A human-approval control plane for AI agents. Catch risky actions — refunds, deploys, emails — before they happen. Approve in Slack or on mobile, gated by po...
|
82 % |
|
ROJO · Behavioral security for AI agents
Rojo is behavioral security for AI agents. We drive your running agent with adversarial inputs, record every tool call it actually makes, and hand you the ex...
|
82 % |
|
AgentAction — Control what agents do. Prove what happened.
AgentAction evaluates agent decisions, enforces policy before actions execute, and checks outcomes against the intended job.
|
82 % |
|
AgentUptime | Les avis de votre riad, avec attention
Suivi des avis Google et réponses personnalisées pour les riads. Français, anglais et arabe. Découvrez deux exemples gratuits avec Mohammed.
|
82 % |
|
Veractly — Let agents act. Know the verified outcome.
Verify consequential agent actions across business systems. Apply policy, inspect the evidence, and recover when the external outcome diverges across your st...
|
82 % |
|
Struxum · AI Action Reliability
Struxum is the reliability layer between AI agents and the external systems they change, making consequential actions safe to execute, retry, verify and reco...
|
82 % |
|
CommitLayer — The evidence layer for AI-agent actions.
Prove what your AI agents actually did. Independent evidence of what your AI agents actually did. CommitLayer verifies, from your systems of record, that eac...
|
82 % |
|
Nexward: Runtime security for AI agents
Nexward sits between your AI agents and the tools they call. Every action gets a deterministic decision (allow, block, or require human approval) before it r...
|
82 % |
|
Strentis — AI Safety Systems
Strentis builds the governance stack for autonomous AI. agent-immune gives you a kernel-enforced authority boundary for coding agents. agent-contract adds ma...
|
81 % |
|
ActPass: stop risky AI agent actions before they execute
ActPass sits between your AI agents and their tools: every risky action gets a deterministic, signed allow/deny decision and a tamper-evident audit trail.
|
81 % |
Detected social media networks