Public-source research
Review what vendors say publicly, the exact quoted source text, the security requirement each statement may support, and what still needs verification.
Showing 241–260 of 1280 evidence records
LatticeFlow AI claims app-specific evaluations and adversarial testing before launch with continuous monitoring afterward.
combining app-specific evaluations and adversarial testing before launch with continuous production monitoring afterward
LatticeFlow AI materials reviewed did not provide a public claim for model and AI component supply-chain inspection.
LatticeFlow AI materials reviewed did not provide a public claim for inline model, agent, tool, application programming interface (API), or Model Context Protocol (MCP) policy enforcement.
LatticeFlow AI claims evaluation of agent workflows, security controls, tool permissions, reliability, compliance risk, and multi-step behavior.
assessing security controls, tool access permissions, reliability, and compliance risk across multi-step agent behaviors
LatticeFlow AI materials reviewed did not provide a public claim for trust or policy enforcement between agents.
LatticeFlow AI materials reviewed did not provide a public claim for non-human identity, service-account, secret, or workload credential lifecycle controls.
LatticeFlow AI materials reviewed did not provide a public claim for agent registration, delegated authorization, task-scoped access, and revocation.
LatticeFlow AI materials reviewed did not provide a public claim for coding-agent, integrated development environment (IDE), CLI, workstation, skill, hook, or package-action governance.
Runlayer claims discovery of unmanaged agents, MCPs, skills, plugins, and client configurations with visibility into agent sessions.
Surface Shadow AI from unmanaged agents, MCPs, skills, plugins, and client configs
Runlayer materials reviewed did not provide a public claim for embedded AI discovery across the business-application environment.
Runlayer claims centralized monitoring of AI usage, adoption, agent activity, and team-level value.
Monitor AI usage, cost, adoption, and agent activity in one place
Runlayer claims policy over what users and agents may access under identity, budget, OAuth grant, and runtime conditions.
Set what users and agents can access, under which identity, budget, OAuth grant, and runtime conditions.
Runlayer claims pre-action runtime scanning of tool calls, outputs, intent, and sensitive data.
scan tool calls, outputs, intent, and sensitive data before risky actions reach company systems
Runlayer materials reviewed did not provide a public claim for browser or software as a service (SaaS)-session controls for employee AI use.
Runlayer claims runtime security inspection before agent actions reach enterprise systems.
Secure runtime execution
Runlayer materials reviewed did not provide a public claim for enterprise AI governance, risk, approval, and compliance workflows.
Runlayer materials reviewed did not provide a public claim for adversarial testing or release assurance for AI systems.
Runlayer materials reviewed did not provide a public claim for model and AI component supply-chain inspection.
Runlayer claims an approved Model Context Protocol (MCP) catalog and governed Model Context Protocol (MCP) gateway across major AI clients.
serve them through a governed MCP gateway across every major AI client
Runlayer claims visibility, policy, and audit across agent sessions, MCPs, skills, plugins, memory, triggers, and scoped permissions.
full visibility into every agent session