Havuz

Build for agents — from the AI Gateway to the entire application stack.

Future AGI, LLM ve AI agent uygulamalarını değerlendirme, izleme, simüle etme ve koruma işlerini self-host edilebilen tek bir platformda birleştiriyor.

Neden yükseliyor · Konu, agent uygulamalarında evaluation ve observability ihtiyacının artmasıyla örtüşüyor; 157 günlük repo yaşına karşılık 2.081 yıldız ve son 28 günde 597 commit geliştirme temposuna işaret ediyor. Farklı yazar sayısı 6; bu veriler ilginin nedenini tek başına kanıtlamaz ve yıldızlar benimseme ya da kalite ölçüsü değildir.

Open-source, local-first AI coding agent usage and cost dashboard for Claude Code, Codex, Cursor, OpenCode, Kiro, and Pi.

ThinkWatch Lite, AI istemcilerini yerel bir gateway üzerinden farklı upstream ve modellere bağlayıp istek yönlendirme, maliyet takibi ve güvenlik kontrolleri sunan masaüstü uygulamasıdır.

Neden yükseliyor · Repo 21 günlükken 789 yıldıza ulaşmış; tek noktadan upstream değiştirme, maliyet ve istek izleme ile relay kaynaklı risklere karşı kontrolleri aynı üründe sunması dikkat çekmiş olabilir. Ancak son 28 gün commit ve farklı yazar verisi ölçülmediğinden ilginin kaynağı ve geliştirme temposu doğrulanamıyor.

Your agent’s trace says “success” but the answer is wrong. Tracedown finds the step that caused it and replays the run with a fix.

FailproofAI, farklı AI agent harness’lerindeki çalıştırmaları izleyen ve tehlikeli tool çağrılarını yürütülmeden önce policy ile engelleyebilen bir araçtır.

Neden yükseliyor · 175 günlük bir repo için 5.166 yıldız ve son 28 günde 84 commit ile 8 farklı yazar, hem belirgin ilgiye hem de yakın dönem geliştirme faaliyetine işaret ediyor. Çok sayıda harness için ortak observability ve enforcement sunma iddiası, agent araçlarının çeşitlendiği dönemde dikkat çekici; yıldızlar tek başına benimseme veya kalite kanıtı değil.

Local-first telemetry collector for AI coding agents — unified OpenTelemetry events for Claude Code, Codex, Cursor and more. Token usage, co

A Claude Code mod that puts a live agent dashboard in your terminal: context and cost, advisor timeline, every permission check, subagent ca

Trace-native CI/CD for AI agents: grade production traces, cluster failures, freeze them into replayable regression tests, and block PRs tha

RL observability for LLM agents: behavior maps, native trainer metrics, rollout investigations, and Python/SQL analysis.

Self-hosted AI API and MCP gateway for organizations: SSO and RBAC, per-user identity for MCP tool calls, PII redaction and tool-call guards

DataBuff is an AI-native APM built on Opentelemetry，with multi-agent troubleshooting out of the box.

OrcaReplay, AI agent çalıştırmalarını kaydedip çevrimdışı tekrar oynatmaya ve bir adımdan farklı bir modelle devam ettirmeye yarayan bir debugging aracıdır.

Neden yükseliyor · 29 günlük bir repo için 171 commit ve 5 farklı yazar kayda değer geliştirme etkinliği gösteriyor; 268 yıldız da erken dönem ilgisine işaret ediyor. Yükselişin belirli bir duyuruya veya kullanıcı benimsemesine bağlı olup olmadığı verilen verilerden anlaşılamıyor.

Local-first MCP tools and dashboard for investigating Codex token usage, credits, costs, caching, and thread patterns.

Self-hosted, Docker-first OpenAI-compatible LLM gateway with provider routing, fallbacks, API keys, and usage controls.

Local-first observability your agent reads back. TMA1 records every LLM call, then routes what it sees into the agent's next turn via hooks

the agent experience layer · observability + memory for AI coding agents (Claude Code + Codex) · local-first, typed, yours

Local LLM cost-tracking proxy for OpenAI, Anthropic, Gemini, and pinned OpenRouter calls with token usage, failure, and billing-integrity re

Agent and LLM API performance monitoring via network packet probe. Measures performance of OpenClaw, Claude, Codex, DeepAgents and more — de

Continuum — the agent runtime by ShyftLabs. Build, orchestrate, ship.

Visualize Pydantic AI agent workflows from Logfire traces as an interactive HTML diagram — tools, nested sub-agents, tokens and exact cost.

Local-first flight recorder for coding agents : replay every run as a live session map, score the context bill, and write the fix back into

The runtime reliability layer for AI agents.

Security-oriented three-axis observability for sandboxed AI agents: model intent, app context, and runtime telemetry into verifiable evidenc

End-to-end Langfuse workshop using a TypeScript Agent to teach the AI engineering loop: tracing, prompt management, monitoring, datasets, ex

Grade every agent, scraper and model in a claim's chain — provenance, trust scoring and audit evidence for LLM pipelines

Local-first desktop control center for OpenAI Codex — sessions, models, tokens, accounts, providers, and AGENTS.md.

Specification-driven multi-agent systems with OpenTelemetry observability, agent traces, benchmarking, MCP, and A2A.

htop for AI coding agents — monitor token usage, costs, and workflows across Claude Code, Cursor, Kiro, Codex, and Copilot

What Claude Code is doing between your prompt and its answer, drawn live from the logs it already writes: every model call, every tool, each

AI Firewall & LLM security toolkit - protect your AI applications from prompt injection, jailbreaks, PII leakage, and adversarial attacks

Phoenix-native LLM evaluation, prompt testing, model comparison, red teaming, and observability for Elixir

Local-first AI Agent gateway & observability for Codex, Claude Code，Dsh and more. 统一模型管理、调用追踪与 Token 成本分析。

Langfuse on GreptimeDB. Self-hosted LLM observability.

OpenTelemetry tracing for DeepSeek Harness (dsh): turns each agent turn into a GenAI span tree — steps, LLM calls with TTFT, tool executions

The agentic debugger for Claude Code, Codex CLI and Gemini CLI: a token usage, cost and context-window tracker that shows what the tokens AR

An analytics tool to explore Claude code usage patterns, errors, and token costs

Open-source middleware for AI agents — catches mid-run failures,compresses stale context, and grows a reasoning library across runs. Any mod

Visual debugging, execution traces, and observability for AI agents.

Deterministic, local-only audit reports for Claude Code AI agent sessions. Rust + SQLite. Zero network calls. MCP server for agents.

TypeScript SDK for the NitroPing audit log — batching client, error capture, Hono/Express/Nitro adapters, and GenAI cost tracing.

Production-grade Node.js RAG system with hybrid retrieval, pluggable adapters, and OpenTelemetry tracing

Self-hosted evidence store for coding agents.

A typed, replayable, budget-aware programming language for LLM agents. Compiles to Python & TypeScript.

Open-source agent reliability platform for discovering, investigating, and fixing production agent failures

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.

Live per-node GPU, host, and vLLM stats for DGX Spark clusters, in your terminal

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

The whole eve stack. On your own machine. Durable Postgres sessions, a Docker sandbox, memory, approvals, schedules, 1,070 tool integrations

Local, self-hosted agent infrastructure — one endpoint for every AI client (Claude Code, Codex CLI, Gemini CLI, OpenCode). Unified model acc