#observability

Mindlas catches your coding agent drifting before the bad code lands. Real-time tracking and correction for context rot, unverified done claims, patch sprawl, and tool loops in your Claude Code session.

Visualize Claude Code dynamic workflows from your terminal — a fan-out→barrier→reduce DAG, a gantt timeline, live concurrency, and a self-contained interactive HTML report. Zero runtime deps.

See where your AI assistant burns money — and pay less. Local profiler for Claude Code, OpenClaw & Hermes Agent: who spends (cron/chats/helpers), the overnight bill, why (loops, retries, idle wake-ups), ready config fixes, world model trends. Zero deps, nothing leaves your machine.

git diff for your agent's context window. See exactly what your LLM saw — turn by turn, block by block.

Lightweight self-hosted NVIDIA GPU fleet monitoring with read-only agents, secure ingestion, compressed storage, and a responsive React dashboard.

Native Windows tray monitor for Claude Code usage — shows your rate-limit %, projects when you'll run out, and turns the icon red before you hit the limit. Plus a local 24h usage breakdown.

Local-first autonomous agent: Hermes speed plus Rust determinism, gated actions, checkpoints, evidence receipts, MCP, skills, and multi-model freedom.

See where LLM money goes, why, and how to cut waste. Local-first FinOps + spend intelligence for Claude Code & the Anthropic API. Pure stdlib, zero deps.

Local replay debugger for Browser Use failures with screenshots, model I/O, failed-step timelines, and public-safe HTML exports.

Precision for your AI spend — see exactly where your Claude Code money goes. Local dashboard + optimization skill (caliper.run)

Ten composable AI skills covering the full arc from idea to launch. Discovery hub for prd-ready, architecture-ready, roadmap-ready, stack-ready, repo-ready, production-ready, deploy-ready, observe-ready, launch-ready, harden-ready. Plug into Claude Code, Codex, Cursor, Windsurf.

Evidence, governance, and static reports for agentic runs

Run-scoped LLM observability control plane backed by LiteLLM

Live view of Claude Code agents and subagents in VS Code: the tree, per-agent context and token spend. Local only, no network.

Zero-config, OTLP-native flight recorder for your backend. Records every request, query, job, and exception — correlated — into a beautiful TUI, a JSON CLI, and an MCP server your coding agent can drive. Local-first.

Tamper-evident audit logging for OpenClaw and other local AI agents: hash-chained, locally verifiable, zero cloud dependency.

Multi-Agent AI Orchestrator 2026 🚀 | YAML, 6+ LLM Providers, ReAct & Swarm

Open standard for AI agent capabilities. Write once, run in any compatible runtime. — pip install intentos

Scenery — one CLI tool for building, running, inspecting, and debugging Go backends. Local-first runtime, APIs, logs, traces, DB tools, generated clients, and AI agent skills included.

Self-healing RAG on n8n with LLM-as-judge eval, drift detection, investigator agent, and canary A/B deploys.

Case study: unified observability + evaluation for automation and conversational AI agents. Versioned error catalog, multi-signal confidence, cost meters.

An observable local Agent loop with controlled tools, permission gates, trace viewing, and history replay.

Local-first OpenTelemetry and Grafana diagnostic kit for Codex, with Loki logs, Tempo traces, Prometheus spanmetrics, privacy-safe derived evidence, and focused Grafana dashboards.

Spring AI chat app that queries Grafana over MCP and exports its own telemetry to the official OpenTelemetry Collector. Notes on where Spring AI diverges from the GenAI semantic conventions.

Local-first VSCode HUD for Claude Code — multi-period token+cost rollups, cross-project session watchtower, Mac/Obsidian integrations, 30+ customizable widgets. Read-only. 100% local. dashable.claude-code-cockpit

Local-first macOS observability for coding-agent context

Autonomous Claude Code CI agent in a rootless Podman sandbox — GitLab CI & GitHub Actions, two personalities (read-write Agent, read-only Advisor), OpenTelemetry audit to Elastic, Helm chart for AKS/OpenShift.

TokenFuse — runtime control for AI agents: per-run budgets, loop detection, burn forecast, kill-switch. Observability shows the fire; TokenFuse is the automatic extinguisher.

Kaleidoskop — replay your baro/Mozaik agent runs visually. Audit log → hexagonal neural firing in your browser.

Cross-platform AI Agent action tracker — log, search, and replay what your AI agents did across all your tools

Detect when Claude silently falls back from Fable 5 to Opus 4.8, log it to a local tamper-evident ledger, and keep your work on the frontier model.

Query billions of AI events by what they mean. PrismDB is a semantic event store: columnar analytics, similarity search, and meaning-based clustering over your LLM and agent telemetry — in one engine, one SQL query, at full retention.

Capability Host Protocol — make agent and tool execution visible, replayable, and ready for governance.

Production-grade observability for Fabric Data Agent (FDA) NL-to-DAX interactions, backed by a Fabric Eventhouse (KQL) database.

Runtime infrastructure for AI agents — lifecycle, supervision, secrets, and deterministic time-travel replay. Built in Rust.

Production-oriented Kimi K3 inference control plane with checkpoint release gates, MoE capacity planning, OpenAI gateway, NVL72 deployment, benchmarks, and observability.

Kubernetes observability suite: a pod/container metadata service plus a per-node agent that tails container logs, scrapes Prometheus targets, cadvisor, node metrics and journald, enriches everything with pod metadata, and exports OTLP

Local-first AgentOps flight recorder and capability firewall for MCP-based coding agents with OpenTelemetry traces.

Enterprise financial intelligence AI agent platform with RAG, SEC EDGAR ingestion, FRED macro analysis, SQL analytics, LangGraph orchestration, security guardrails, evaluation, and observability.

Enterprise Agentic AI Platform for autonomous pipeline monitoring, root cause analysis, incident retrieval, data quality validation, and self-healing workflow orchestration using LangGraph, FAISS, Vector Search, and Python.