TraceRoot
An agent run can finish without errors and still give the wrong answer, so TraceRoot traces every model and tool call, flags the production runs that missed the mark, and helps your coding agent ship verified fixes. Each run becomes a browsable call tree with inputs, outputs, latency, and cost, searchable by session or filter. Instrument with the Python or TypeScript SDK, the agent-native CLI, or any OpenTelemetry exporter, with ready integrations for LangChain, LangGraph, OpenAI Agents SDK, CrewAI, Vercel AI SDK, Mastra, and Pydantic AI. Detectors use LLM judges to score live traffic against the behavior you define and can post alerts to Slack. Judge outputs cluster into signals, then a background coding agent reads the traces and your connected GitHub code in a sandbox and recommends changes down to the commit, file, and line. Datasets, test cases, and scored eval runs compare versions so regressions surface before a release, and dashboards of widgets chart your trace data. The CLI hands your coding agent, such as Claude Code or Codex, the traces, signals, detectors, and evals to work from. Self-hosting keeps all of that trace data on infrastructure you control. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed (core edition).
LangWatch
Every call your AI agents make becomes a searchable trace in LangWatch, and the same platform tests and scores those agents before a new prompt or model reaches users. The Trace Explorer filters by origin, user, model, cost, and evaluator, reads each run as a waterfall, flame graph, topology, or sequence diagram, and takes traces from the Python, TypeScript, and Go SDKs or any OpenTelemetry source. Agent Testing sends simulated users, in text or beta voice calls, through your agent while a judge passes or fails every criterion with its reasoning, adds multi-turn red teaming, and compares two agents side by side. The Experiments workbench scores prompts and agents on datasets with LLM judges, exact match, and RAG checks, then compares runs on cost, latency, and pass rate, while online evaluations monitor live traffic and guardrails block unsafe replies. Prompts keep version history and staging or production tags, Workflows wire prompts, Python code, and HTTP calls on a visual canvas, and annotation queues route traces to human reviewers. Alerts fire when a graph crosses a threshold, automations post to Slack, email, or webhooks, and scheduled reports go out by email or Slack. Claude Code, Codex, Copilot, Gemini CLI, and opencode report session cost, pull request cost, and terminal replays, and an MCP server exposes trace search and prompt management to coding assistants. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.
SigNoz
With over 31,000 GitHub stars and native OpenTelemetry support that eliminates vendor lock-in from day one, SigNoz delivers full-stack observability covering metrics, traces, and logs in a single pane of glass without the per-host pricing model of commercial APM platforms. The platform ingests telemetry data through the OpenTelemetry Collector, supporting auto-instrumentation for Java, Python, Node.js, Go, Ruby, PHP, and .NET applications with zero code changes required for basic tracing. ClickHouse serves as the columnar storage backend, providing fast aggregation queries over billions of spans and log lines with configurable retention policies and tiered storage. The distributed tracing view renders flame graphs and Gantt charts showing request flow across microservices with latency breakdowns, error rates, and p99 percentile calculations. Custom dashboards support PromQL and ClickHouse SQL queries with time-series charts, bar graphs, tables, and value widgets. The log management pipeline supports structured and unstructured logs with full-text search, log pipelines for parsing and enrichment, and correlation with traces via trace IDs. Alert rules can be configured on any metric or log query with notification channels including Slack, PagerDuty, OpsGenie, webhooks, and email. The exceptions monitoring module automatically groups and tracks application errors with stack traces, occurrence counts, and first-seen timestamps. Service maps visualize inter-service dependencies with real-time latency and error rate overlays. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed with an enterprise edition available.
Monoscope
Across logs, traces, and metrics, Monoscope gives engineering teams one self-hosted workspace for finding what broke, asking questions in plain English, and hearing about problems before users do. Start in the Explorer: search events with KQL or an Ask AI prompt, flip between log rows, bar and line charts, log patterns, and sessions, and narrow results with facets for service, level, and HTTP status. Open any span to read its attributes, then follow the full trace across services in a waterfall, timeline, or map view. Live Tail streams new events as they arrive, and the Service Map draws every dependency with its error rate, latency, and requests per second. Build dashboards from widgets with copyable KQL and SQL, then set scheduled monitors with warn and alert thresholds, cron monitors, and uptime checks. An Issues inbox gathers API changes, errors, and log patterns so each one can be acknowledged, archived, or marked as spam, while Reports send scheduled summaries by email. Telemetry arrives over OTLP gRPC on port 4317 from OpenTelemetry SDKs and collectors. The monoscope CLI emits JSON for scripts, and an MCP server at /api/v1/mcp lets Claude, Cursor, and Cline search events and triage issues. Natural-language search uses the OpenAI-compatible model you configure. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.
Apache SkyWalking
An open-source application performance monitoring platform for microservices, Kubernetes, and service meshes, Apache SkyWalking draws the live service map and ties slow requests to their traces, logs, and metrics. Agents cover Java, .NET Core, PHP, Node.js, Go, Lua, Rust, C++, browser JavaScript, and Python, and OpenTelemetry, Zipkin, Prometheus, Zabbix, and Telegraf data can be ingested too. Start at the service topology map, follow a request through the trace waterfall, then drill into the instance and its API dependencies. When the cause is inside the code, trace profiling, async-profiler, Go pprof, eBPF, and network profiling render flame graphs and process graphs in the same console. The Rover eBPF agent profiles Kubernetes workloads for CPU and network problems, and Istio and Envoy service meshes, including Istio Ambient, report into the same dashboards. Alarm rules target a service, deployment, or API, and alarms and telemetry can be forwarded to third-party systems. Metric and log analysis rules hot-update at runtime without a restart, and a live debugger shows what each rule stage receives and produces. An AI assistant answers plain-language questions with charts, topology, and traces from live data, using an OpenAI-compatible or Amazon Bedrock model you configure, and an MCP connection serves your own AI client. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.
Parseable
Parseable replaces expensive Elasticsearch clusters and fragmented monitoring stacks with a single Rust binary that ingests, queries, and stores logs, metrics, and traces on commodity object storage at a fraction of the cost. The data lake architecture decouples stateless compute from S3-compatible storage, enabling independent scaling of ingestion throughput and query capacity while cutting storage costs by up to 90% compared to indexed alternatives. OpenTelemetry-native OTLP ingestion accepts telemetry from existing OTel collector pipelines, Prometheus Remote Write endpoints, Kafka consumers, eBPF probes, and popular logging agents including Fluentd, Fluent Bit, and Vector without proprietary format conversions. The SQL-first query interface enables cross-signal analysis across all telemetry types, while native PromQL support with 50+ functions and 12 aggregation operators provides Prometheus-compatible metrics querying that works directly with Grafana dashboards. Built-in features include customizable dashboards, real-time alerting with Webhook, Slack, and Alertmanager targets, role-based access control, OpenID single sign-on integration, LogIQ automatic unstructured-to-structured log transformation, smart caching for frequently accessed data, and retention policies for lifecycle management. AI-powered Keystone Q&A provides natural language to SQL conversion and dataset summarization. All data stored as standard Apache Parquet on object storage remains accessible to any Parquet-compatible engine (DuckDB, Spark, Trino), ensuring zero vendor lock-in. Deploys on AWS S3, Azure Blob, Google Cloud Storage, MinIO, Wasabi, and DigitalOcean Spaces. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.
ClawMetry
Without an SDK, a proxy, or a single line of agent code changed, ClawMetry reads the session files your AI agents already write and puts every session, tool call, token, and dollar on one live dashboard. The sessions view replays a run turn by turn, with a timeline that gives each tool call its own duration and cost, so a four-minute shell command stands apart from a call that took milliseconds. Usage charts split tokens and spend by runtime, model, session, and day. A live flow diagram follows messages from chat channels through the gateway to models and tools, and a context view tracks window use per turn along with compaction events. The Brain tab streams reasoning, tool calls, and errors as they happen, while memory, skills, health, logs, and cron jobs each get a tab of their own. OTLP receivers accept traces, metrics, and logs from runtimes that keep no local files. The MIT-licensed app reads OpenClaw, NVIDIA NemoClaw, Goose, and Qwen Code for free. Paid plans add the other 28 runtimes, including Claude Code, Codex, and Cursor, plus approval queues for risky tool calls, budget limits, anomaly detection, and a cost optimizer. Session data stays on your server unless you choose to run cloud sync. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Rootprint
A year of searchable logs and traces no longer needs a metered logging service, because Rootprint keeps the whole index in a storage bucket you already own. A search matches any word on any line, since every log is indexed in full and no service name, stream, or label is needed first. A frequency histogram, one-click field filters, and saved views narrow the results, a fold mode collapses repeated lines, and exports arrive as NDJSON, CSV, or plain text, or as a share link. Open a trace from the log that emitted it, or paste a trace ID, and a waterfall shows per-span timing, attributes, events, and database calls grouped by query, with each span linking back to its logs. The trace explorer charts span volume, error rate, and p50, p95, and p99 latency, ranks the busiest operations, and filters spans by service, duration, and status. Every service gets its own page with request rate, error rate, operations, dependencies, and errors. Data arrives over OpenTelemetry from the Collector, Vector, or Fluent Bit, through a plain HTTP gateway, or by pulling from Kafka, Kinesis, or SQS, with guided setup for nine integrations. Indexes can live in S3, Google Cloud Storage, R2, MinIO, or Azure Blob. Teammates sign in with Google, GitHub, or any OpenID Connect provider, and scoped API keys serve scripts. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.