Havuz

A harness for hosted Grok Bot crews. Its core rule: it is not allowed to guess.

A token-spend profiler and cost-regression gate for AI agents.

Uncover the hidden world of AI agents & MCP servers 🐙

Watch your Claude Code subagents work as a pixel-art office. Rewind any second, see what every agent cost, and turn any session into a timel

See your whole Kubernetes cluster at a glance. Lightweight, web-based, real-time, single Go binary visualizer.

Langfuse-style trace viewer for pi-agent — local-first, zero-setup, single-file HTML

Distributed, centrally-controlled Web Application Firewall built on pingap & Cloudflare Pingora. Semantic attack detection, Cloudflare-style

Prometheus cloud SD service for discovering AWS and Alibaba Cloud databases, middleware, and node exporter targets.

Diff your AI agent's behavior between two runs. See exactly which tool calls, args, costs and outputs changed when you swap models or edit p

eBPF egress firewall for Kubernetes — see and enforce what every pod talks to (domains, HTTP paths, headers, IPs), attributed per pod, enfor

Graph-native studio, runtime integration, and self-hosted observability for mutable multi-agent LLM workflows.

Open-source, self-hosted analytics for MCP servers

Semantic cache strategy bench: replay recorded request traces through candidate caching policies and measure which one actually pays off.

Live code graph that fuses static analysis (tree-sitter) with OpenTelemetry runtime traces, queried over MCP so AI coding agents get real fu

Universal Bash watchdog for monitoring websites, HTTP endpoints, TCP ports, and command-based services. Configure checks, retries, timeouts,

Open-source eBPF compatibility evidence and CI gate

Hands-on GPU/HPC infrastructure operations: K8s GPU scheduling, HAMi sharing, Slurm, observability & vLLM inference. Learn it free on a lapt

Local-first observability and evaluation for production forecasts - capture, validate, and compare forecasts with a local UI.

Evidence-grounded evaluation for AI agents — verifies each claim against the agent's real tool outputs (constrained, evidence-grounded model

Deterministic cost / loop / time budgets · full observability · crash-resumable runs · human-approval gates · a memory you own. Self-hosted.

Lightweight observability for multi-agent AI pipelines. Self-hosted, no cloud dependency — add two lines of Python and see real-time agent e

Tailscale exporter for OpenTelemetry and Prometheus. Device fleet, network flow logs and audit logs as OTLP metrics & logs. Single Go binary

A public repository dedicated to helping organizations, students, and IT professionals learn, deploy, and operate a production-ready, open-s

Codex Usage — Local-first Codex usage analytics · 逐电脑/模型/项目/会话 Token 用量明细与 API 等价成本，支持 Windows/Linux/Mac。

MCP server for Asterisk and FreeSWITCH. Lets AI assistants inspect channels, SIP registrations, trunk status and dialplan on a live PBX.

Hound is a Kubernetes network observability agent that exposes TCP connection-level metrics (state, throughput, retransmits, RTT, SYN anomal

MLOps — vLLM + LiteLLM + Portainer CE + Netdata. Case study song ngữ: kiến trúc, metrics, bài học vận hành. · Self-hosted LLM inference don

See where your Claude Code tokens go — cost per call scales 4.3x with context size. Hooks + SQLite + dashboard + MCP server for context wind

Rust based Postgres extension that adds query lifecycle events to OTEL traces.

Tree-shakable health endpoints for JS servers. Zero-dependency core, adapters for Hono, Elysia, Express, Next.js and TanStack Start, probes

Thermal guard for Apple Silicon: pauses hot jobs with SIGSTOP and resumes them cool, optional admission queue against a thermal budget, keep

Session replay and distributed tracing on one timeline - watch the user's session and the backend spans behind it.

Fault injection framework for on-device AI. Pure Dart core (sate_core) + Flutter package (sate_ai). Android, iOS, Web, and CLI targets.

nvtop for vLLM — an interactive terminal dashboard for vLLM serving performance (concurrency, throughput, cache & KV memory, latency, spec-d

CLI-first, local LLM cost tracker for indie devs — drop it in front of any OpenAI-compatible API and watch token usage + spend in real time.

A Symfony Web Profiler-inspired toolkit for NestJS — profile every request with a rich panel UI. Powered & maintained by Eleven Labs.

A local-first, MCP-native toolkit for building AI agents: engram (memory) + skillet (skills) + tracelet (tracing). One runnable end-to-end r

Session-layer circuit breaker for LLM & agent traffic — a transparent streaming proxy that meters every session and kills the one that goes

Device management & monitoring system. Go backend (Chi + SQLite, CGO-free) with an embedded SvelteKit 5 SPA — single-binary deploy. SNMP/ICM

Ground truth for enterprise AI: your AI coding tools, your accounts and every session on your own server, from one web console, CLI or API.

Self-hosted error tracking. A drop-in for Sentry and AppSignal SDKs: one Go binary and Postgres, with Slack, Linear, email and an MCP server

The Platform Engineering CLI. Generate production-ready modern JS/TS architectures in seconds (starting with Next.js). Features built-in AST

One npm package. Durable workflows, feature flags, auth, request tracing, rate limiting, caching, metrics, webhooks, CRON, and more — all se

Realtime dashboard for the Halogen LLM inference server

See which notes your coding agent actually reads — a live radar for Claude Code and your Obsidian vault.

Self-hosted OTLP receiver and dashboard for traces, logs, and metrics. .NET 10 + Nuxt 4 SPA, with SQLite, PostgreSQL, or SQL Server storage.

Welcome to cerbere-agent guard, originally just agent guard, then rebranded after the mythical hound of the underworld. Our mission? Guardin

DSH 插件：统计所有经过 DeepSeek Harness 的模型调用——Token 用量、成本核算（含峰谷计价）、供应商余额与配额，并按模型/供应商/项目/来源/账号/日期拆分。A token usage, cost and provider-balance plugin f

An AI agent that load-tests a code change, finds the regression in Splunk, and writes the fix. A model-agnostic, audited state machine over

OpenTelemetry tracing for pi-coding-agent — per-turn span tree with full OTel GenAI semantic conventions