Search

Local-first AI workbench for llama.cpp models. Full chat UI with SSE streaming, multi-engine web search, sandboxed Python, self-aware source tools with diff previews, hot-loadable plugins, TaskMaster prompt decomposition, two-tier wiki RAG, and inline Chart.js/SVG applets. Runs entirely on your hardware.

Graph-first agent framework on the BEAM (Elixir/OTP) — durable execution, HITL, time-travel, checkpoints, MCP. A LangGraph alternative, no Python.

Self-hosted orchestration daemon: drives Codex CLI and Claude Code agents across a Linear backlog in isolated per-issue workspaces, with a Phoenix LiveView observability dashboard.

Voice Agent implementations for the VAmoS Benchmark

Formally verified mathematical foundations for multi-agent AI coordination

Tools and Functions that I use regularly in my agents

AI assistant framework core handling agent sessions, plugins, and tools.

Scaffold production-grade multi-agent AI systems into any repository

One client for everything a Workser project owns — database, business data, connected apps, storage, auth and workflow automation. Zero-config inside a Workser project.

Record a browser walkthrough as a semantic event stream (JSONL) — multiple candidate selectors, recorded assertions, and intent notes — then let an AI agent convert it into tests that reuse your existing step library.

AI Agent Cost Guardrails — budget enforcement for LLM calls. Check before, record after.

Plataforma visual self-hosted para agentes e modelos locais de IA

Agent observability for Python. Trace LLM calls, track cost/latency/tokens, export to Qhaway HTTP API, OTEL, or local SQLite.

Minimal TypeScript SDK to orchestrate AI agents and tools — a lightweight, transparent alternative to LangChain.

A local agent daemon core built around a usable agent harness and domain events.

Unified observability for AI agents and infrastructure. One SDK.

Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.

AgenticRail gate client for JavaScript and TypeScript

Temporal Integrity Probe — behavioural consistency testing for LLM agents under temporal context manipulation (GenesisAeon P50)

Pure-Python Multi-Agent Reinforcement Learning Engine