runtime-safety

Deterministic safety solutions for probabilistic AI agents

🛡️ Zero-dependency runtime crash guard for Node.js apps with disciplined exits and clean logs

Runtime governance for AI agents. Contracts enforce what tools can do — before they execute.

cppnat is a NAT reverse proxy tool (aslo known as a hole punching tool) helps you expose your private IP and port and bind them onto a server with public IP,access your private service anywhere.

a library aims to convert json string to C++ data struct in modern,fluent and fast way. Runtime safety and No C++ Exception.

Local-first runtime safety layer for AI agents that blocks runaway costs, loops, retries, and budget overruns before provider API calls execute.

AI Identity and Behavioural Fingerprinting

A modern,fluent and simple C++ mysql database library implemented wit C MySQL API.Runtime safety and no exception.

Runtime guardrails for AI agents that enforce token budgets, loop limits, and tool rate limits locally.

Runtime approval gates for AI agent tool calls. Intercept payments and emails before execution

Ts-TypeSafeEnv: A lightweight TypeScript library for managing type-safe environment variables. Say goodbye to runtime errors with compile-time validation and effortless setup—perfect for building reliable, error-free JavaScript or TypeScript projects 🚀.

Runtime safety components for AI model inference with formal proofs, ultra-low latency, and guaranteed correctness.

Verified maintenance infrastructure for coding agents: detect upstream changes, generate isolated patches, and prove the exact change before it ships.

AgentShield — trust layer for AI agents. Risk scoring, pause-for-approval gate, and kill switch for autonomous AI systems.

Stop writing schemas twice. Define logic as functions, let Typebox infer the types and activate them into a callable facade.

Detect the most common JavaScript runtime crash before it happens

Runtime policy guard for AI agent processes: watches process, file, and network activity from user space, applies deterministic policy, and attempts SIGSTOP or SIGKILL on reviewed triggers — no SDK required, works against any PID. Audit-only mode rehearses policy first; every decision logged as evidence.

Runtime safety through convex projection

A lightweight safety primitive for multi-channel estimator disagreement in autonomous systems. Triggers staged, reversible, low-risk behavior until estimates reconverge. Research prototype — not certified.

Modular middleware for emotional safety, symbolic transparency, and real-time drift prevention. From bind-aware routing to EU-ready ethics, LLOYD Preserver protects behavior, motifs, and trust in language systems.