TypeScript client for the Vaara HTTP API: EU AI Act runtime evidence for MCP tool calls. Conformal risk scoring, policy gating, hash-chained tamper-evident audit, named detectors.
Open-source command center for governed AI coding agents with built-in onboarding, hard gates, MCP, and shareable run receipts.
A timelapse of every request — full request stack trace (traceId + before/after of each table change) for Node.js.
Composite record + blob archive export for noy-db — bundles a collection's records plus every attached blob into one zip. Zero dependencies (STORE-only zip encoder). Gated by the RFC #249 canExportPlaintext capability bit with format 'zip' . Part of th
AI agent governance - audit trails, policy enforcement, ML-DSA cryptographic signatures
Governance layer for multi-agent AI systems. 7 mechanically enforced rules, 30 tools including kill switch, audit chain, dispatch, drift audit, security audit, page health, pre-publish audit, and session close. Works with Ruflo, Hivemind, Agent Teams, and
Network control plane for loom — an HTTP transport over the headless driver-loop and a registry that supervises a fleet of projects from one process. Submit a task, read status, and answer a gate over localhost; intake adapters (a dashboard, a chat bot, a
MCP server for AI agent authorization — register agents, issue scoped tokens, verify grants, audit actions, revoke access
Configurable audit logging for Drizzle ORM
Anchor-at-write-time for the ar.io verification stack. Hash your data locally, sign a Verifiable Event Envelope (ario.events/v1), and anchor it to ar.io — permanent, tamper-evident provenance with Merkle-batched checkpoints and per-event inclusion proofs.
Core types, schemas, and utilities for AgentLens — AI agent observability platform
MCP transport for loom — exposes the pipeline tools over stdio and assembles the runnable registry.
AgentLens API server — event ingestion, querying, dashboard, and alerting for AI agents
Local-first AI coding navigation CLI and MCP server with step gates, audit trail, and SOP-aware project state.
Transport-neutral orchestration runtime for loom — the headless driver-loop, the directive→wire adapter, and the conformant delivery composition shared by every transport.
Anchor-as-you-store for S3: wrap your S3 client so every putObject also anchors a tamper-evident provenance record to ar.io and writes it beside the object as a verify-anywhere .provenance.json sidecar. Bytes are hashed locally, never uploaded.
Official Palveron TypeScript/JavaScript SDK — AI Governance Gateway client for policy enforcement, trace verification, and compliance automation.
Long-lived supervisor over loom's headless driver-loop — drives a project's task server-side, parks on a human gate and wakes on the answer, retries transient failures, recovers in-flight work on restart, and owns the worktree lifecycle.
loom's persistent control layer — the global config + secrets + project catalog, and the one layered resolver every face (CLI today, server/UI later) reads. A leaf: it depends on nothing else in loom and learns no domain. Configure keys/models once; inher
AgentLens API server — event ingestion, querying, dashboard, and alerting for AI agents
Database-backed state machine for Node and TypeScript — core (ORM-agnostic).
Prisma + Postgres adapter for stateledger — persisted, audited, concurrency-safe state machines.
Reference implementation of the Ironclad standard — multi-agent dev harness for Claude Code.