Real-time tracing, replaying, and rewinding of AI coding assistant edits. MCP server for AI self-correction. Cross-agent support (Claude Code, Cursor, Codex). Export to Agent Trace JSON and git-ai git notes.
A high-performance CLI tool for LLM prompt regression testing
Constant-time-designed pure-Rust SM2/SM3/SM4 primitives (no_std + alloc) with an in-CI dudect timing-leak regression harness
SIMD backends for gmcrypto-core — AVX2 (x86_64) and NEON (aarch64) packed bitsliced SM4 S-box, quarantined to keep `gmcrypto-core` `unsafe_code = forbid`
API deprecation management agent for Sentinel proxy - sunset headers, usage tracking, and migration support
runtime defense for CLI AI agents — intercepts tool calls before execution and enforces security policy
This is Pre alpha, and not useful, no code worth looking at.
AI Gateway agent for Sentinel reverse proxy - prompt filtering, PII detection, usage control
Prism — Reliability layer for agentic AI systems: voting mesh, checkpoint/replay, and Python bindings
Static security analysis CLI for Solana Anchor programs
Dependency upgrade impact analyzer — test upgrades against your projects' real CI before you adopt them
Three‑pane terminal file viewer (TUI) with preview and Lua configuration
This gem enables access authorization control, based on the access permission settings for each controller and its actions, at the model level.
Group access authorization control, based on controllers actions access permissions using ActionSentinel.