observability

Open-source LLM gateway that adds cost visibility, request tracing, repeated-call detection, and basic prompt-injection heuristics for OpenAI and Anthropic traffic.

Marginal is an LLM cost-tracking and observability product that attributes spend by customer, feature, model, or any custom field. The page shows a dashboard, pricing logic, SDKs, HTTP API access, and example integrations for TypeScript and Python.

An AI incident copilot that watches your screen during an incident, extracts what happened, and answers follow-up questions using the Grafana LGTM stack.

Local-first observability and evaluation layer for production forecasts. Captures forecasts, validates them, computes horizon-aware metrics, and serves a read-only local UI for comparing runs.

Self-hosted observability and control plane for production and coding AI agents, with live timelines, fleet-wide feeds, token budgets, MCP allow/block rules, and support for Claude Code plus a Python sensor.

Open Bias is an open-source reliability harness for LLM apps and agents that enforces rules at runtime. It sits between your app and an LLM provider, using RULES.md policies to trace, block, or fix off-policy behavior.

Open-source runtime guardrails for AI agents with cost limits, loop detection, policy enforcement, observability, and dashboard/CLI support.

Telemetry is an observability platform for AI and LLM apps that tracks model calls, tool steps, retrievals, tokens, cost, latency, and errors. It supports OpenTelemetry ingest, TypeScript and Python SDKs, and integrations like Vercel AI SDK and LangChain.

DriftWatch AI is a proxy and middleware layer for LLM traffic that trims redundant tokens, validates JSON schemas, and monitors model drift across OpenAI, Anthropic, and Gemini workflows.

Voidleap Code is a local-first agentic coding desktop app for Mac and Windows. It lets developers run AI coding workflows on their own machine, use their own provider keys or local models, switch models mid-task, and inspect prompts, tool use, context, and cost telemetry.

Open-source package collection for BetterDB’s Valkey/Redis-based AI infrastructure, including semantic cache, agent memory, retrieval, monitoring, CLI, and an MCP server for assistants.

Open-source CLI and TUI for auditing Claude Code skills: shows usage heat, token cost, broken skills, and relationship graphs, with JSON/Markdown output for CI.

Open-source runtime for multi-agent AI systems with job scheduling, DAG orchestration, shared memory, tool execution, and real-time observability.

RiskKernel is an open-source, self-hosted reliability runtime for AI agents. It adds hard budgets, kill switches, crash-resumable runs, human approval gates, and OpenTelemetry export, with support for OpenAI, Anthropic, and existing agent stacks.

Intencion is a product analytics and observability platform for AI agents. It tracks user intent, tool calls, run outcomes, and resolution trends, then turns failures or unmet requests into Linear or GitHub issues.

Open-source platform for governing AI agents across clouds and stacks, with a control plane, sidecar-based data plane, policy enforcement, observability, and audit trails.

Greenflash is a product management layer for AI agents that reads production conversations, surfaces friction and intent gaps, and recommends what to fix or ship next. It includes SDKs, a REST API, Claude Code integration, and support for voice and text agents.

Self-hosted runtime for putting hard budgets, approval gates, observability, and resumable checkpoints around AI agents.

A local, read-only observability dashboard for Claude Code with live event streams, session Kanban, token and cost charts, a 3D tool-call graph, and plugin-based widgets.

Open-source agentic mesh for governing AI agents across enterprise systems, with identity, policy enforcement, audit trails, observability, and Kubernetes-native deployment.

An open-source control plane for AI agents that adds policy enforcement, approval gates, durable execution, audit logs, and observability for unattended workflows.

Go-native framework for building, orchestrating, and observing AI agents with native OpenTelemetry, an embedded dashboard, replay, and first-party MCP/A2A support.

An OpenTelemetry Collector distribution that turns GenAI traces into bounded metrics and keyed summaries, designed for fleet-level observability without exporting raw prompts or high-cardinality labels.

A logical KV cache profiler and leak detector for PagedAttention engines like vLLM and SGLang, with a dashboard, live engine hook, CI checks, and Perfetto trace export.