AWS infrastructure tools: EKS, RDS, Lambda, CloudWatch, SQS monitoring + declarative tool engine

Core runtime and API client primitives for rtrvr CLI/SDK

OpenAI-compatible client + agent for Cana Apps — Bearer auth (App API key or short-lived browser JWT with auto-refresh), server-hosted MCP federation, streaming, typed structured output. Node 18+, Bun, Deno, Workers, browsers.

GitHub tools: PRs, commits, reviews via GitHub App installation auth

CLI executor — runs co-located impl.{mjs,py,sh} via stdin/stdout JSON

HTTP executor — runs declarative HTTP tools with per-user auth resolution

Typed TypeScript client for the Koda Rising API — auth, multi-currency wallet & billing (NGN, KES, GHS, ZAR, EGP, USD), the agent (run + streaming run-loop), chat, teams, and org-scoped retrieval.

Jira tools: issue search, creation, comments, transitions via Basic auth

MCP-client executor — delegates tool calls to an upstream MCP server

Familiar plugin: long-term memory for the agent. Search backend pending replacement; memory_save handler still active.

Agent observability SDK for mapping host runtime events into Langfuse traces.

Base types, constants, and utilities for the BoringOS framework

Reusable React components for PAI Agent chat and widgets.

GCP infrastructure tools: GKE, Cloud SQL, Cloud Functions, Cloud Logging, Pub/Sub monitoring

Tiny TypeScript step checkpoint wrapper for resumable agent runs.

Official TypeScript/Node SDK for TrustModel — AI trust evaluation, OTel telemetry, and agent governance (AGP).

Vercel AI SDK adapter for agenttrace-react. Visualise AI SDK agent execution traces, tool calls, and human-in-the-loop approval gates as a live React UI.

GPTAgent.js is a composable and extensible framework for creating AI agents with TypeScript/JavaScript.

Deterministic, durable and tenant-isolated verification control plane for immutable source candidates.

Bidirectional sync adapter — Plane project management integration, conflict detection, and sync state management