WORK WITH BUSINESS CLIENTS?

Sell AI employees under your own brand.

GlassFlow Rius SDK logo

GlassFlow Rius SDK

GlassFlow Rius SDK AI Agent
Rating:
Rate it!

Overview

OpenTelemetry-native SDK for tracing, debugging, and monitoring production LLM agents in GlassFlow Rius.

GlassFlow Rius SDK instruments Python and TypeScript AI agents so teams can send traces, spans, LLM calls, tool activity, token usage, cost data, errors, and heartbeat signals into the managed GlassFlow Rius observability platform. It is designed for teams running short sessions as well as long-running agents that may span hours or days. Developers create a Rius account, generate an API key, install the SDK, initialize it at application startup, and wrap agent functions so traces stream into Rius for live debugging, alerts, workflow maps, dashboards, and MCP-based querying from tools such as Claude Code, Cursor, or ChatGPT Codex. Rius does not replace the agent runtime or make business decisions itself; it gives engineers and AI platform teams visibility into what agents did, where they failed, and what humans may need to fix, approve, or investigate.

Product overview

Capabilities, pricing and fit

GlassFlow Rius SDK is useful for teams that already have LLM agents in development or production and need operational visibility rather than another agent builder. It sends OpenTelemetry-native traces from Python or TypeScript agents into GlassFlow Rius, where teams can inspect live runs, debug failures, monitor cost and tokens, receive alerts, and query trace history through MCP-enabled coding tools.

Information as of November 20, 2026

What it can do

  • Agent trace collection

    Rius traces every LLM call, tool invocation, and agent step so teams can inspect agent runs and failures.[4], [5]

  • OpenTelemetry-based instrumentation

    The SDK uses OpenTelemetry-style instrumentation and can be added with initialization plus observe wrappers for Python or TypeScript agent code.[5], [7]

  • Live debugging and trace visualization

    After an agent run, traces appear in Rius with a waterfall view that includes steps, token counts, cost, and errors.[5]

  • Long-running agent monitoring

    Rius is positioned for both short chat sessions and long-running agents, including heartbeat detection and sessions across restarts and sub-agents.[1], [3]

  • Cost and token tracking

    Rius surfaces token counts and cost for traced agent activity, and listed plans include cost and token tracking.[3], [5]

  • Alerts and dashboards

    Every listed plan includes real-time traces and spans, alerts, dashboards, cost and token tracking, and agent workflow maps.[3]

  • MCP trace querying

    Rius supports MCP workflows so coding agents or editors can query traces, errors, cost, and alerts in natural language.[4], [5], [8]

  • Python and TypeScript SDKs

    The quickstart documents Python and TypeScript installation and examples, including Anthropic auto-instrumentation.[5], [6]

Where it fits best

  • Engineering teams running LLM agents in production that need to see every agent step, LLM call, tool invocation, error, cost, and token signal in a traceable workflow.[4], [5]
  • Teams operating long-running agents that need heartbeat detection, live metrics on running agents, and continuous sessions across restarts or sub-agents.[1], [3]
  • Developers who want to query agent failures and trace data from coding environments through MCP rather than relying only on a dashboard.[1], [4], [5]

Pricing and availability

Rius is priced by usage rather than by seat. Listed monthly plans start with Core at $19/month, Growth at $79/month, Pro at $199/month, and Enterprise at $1999/month. Plans include different span allowances, retention, throughput, users, workspaces, alerts, and support levels. New organizations start with a trial lasting 30 days or 500,000 spans, whichever comes first.[3], [5]

Platforms: Python, TypeScript, OpenTelemetry, MCP-compatible clients, Anthropic instrumentation[3], [4], [5], [6]

Access: Managed Rius cloud account, SDK embedded in an existing agent application, EU/US data regions on listed plans, Dedicated regional cluster add-on[3], [4], [5], [6]

Important considerations
  • Rius requires a GlassFlow Rius account, an API key, and SDK instrumentation before traces appear.[5]
  • The documented new-organization trial is limited to 30 days or 500,000 spans, whichever comes first.[5]
  • Paid usage is priced by trace or span volume rather than by seat, so teams should estimate trace volume before selecting a plan.[3]
  • For short-lived TypeScript scripts, developers must call shutdown before process exit so queued spans are exported.[5]
  • Standard data-region choices are EU/US on listed plans, while selecting a specific regional cluster is a paid add-on or enterprise-style configuration.[3]
Sources (8)
  1. AI Agent Observability & Data Infrastructure | GlassFlowOfficial site · checked 2026-11-20
  2. GlassFlow Rius | AI Agent Observability in ProductionOfficial site · checked 2026-11-20
  3. GlassFlow Rius Pricing | Agent Observability from $0Pricing page · checked 2026-11-20
  4. GlassFlow DocumentationDocumentation · checked 2026-11-20
  5. Quickstart – GlassFlow DocsDocumentation · checked 2026-11-20
  6. Installation and configuration – GlassFlow DocsDocumentation · checked 2026-11-20
  7. GitHub - glassflow/rius-sdk-pythonGitHub · checked 2026-11-20
  8. ai.glassflow/rius - AI Tools MCP Server - MCPKitDocumentation · checked 2026-11-20

Autonomy level

18/100

Our subjective rating of how independently the product carries out its documented tasks, considering the human setup, input, approval and review it needs. This is not a measured success rate.

Reasoning: GlassFlow Rius SDK is an instrumentation and observability layer, not an autonomous agent runtime. After developers add the SDK and API key, it can automatically emit traces, spans, heartbeat signals, LLM call metadata, token counts, cost, and errors to Rius, and the platform supports alerts and MCP queries. However, humans still configure the agen...

Comparisons


Custom Comparisons

Some of the use cases of GlassFlow Rius SDK:

  • tracing production LLM agents
  • debugging agent failures
  • monitoring long-running agents
  • tracking token usage and cost
  • querying traces from coding assistants
  • adding alerts for agent errors and loops

Loading Community Opinions...

Pricing model:

Code access:

Popularity level: 22%

Free credibility widget

Turn this profile into a trust signal

Show prospects that GlassFlow Rius SDK has a public place where they can check product details, pricing, ratings, and reviews.

Build confidence

Give buyers a third-party profile to explore.

Reduce hesitation

Put validation beside your strongest CTA.

Earn discovery

Every badge links prospects to your listing.

Choose your style

Preview it, then copy the complete embed code.

Live previewReady to embed

Shows buyers where to validate your product, pricing, and reputation.

Plain HTML. No signup, script, or maintenance required.

FOR AGENT BUILDERS

Run your agents in a persistent workspace.

Run Hermes and OpenClaw with saved context, connected tools and delegated coordination.

Did you find this page useful?

Not useful
Could be better
Neutral
Useful
Loved it!