Riva Command Center

Local-first observability for the coding agents on your machine.

Claude Code, Codex, Cursor, Copilot — as agents get more autonomous, visibility disappears. Riva restores it: see what every agent is doing, what it costs, and whether it's behaving. All local. Your data never leaves your machine.

macOS + Linux

curl -fsSL https://rivalabs.ai/install.sh | sh

Windows

irm https://rivalabs.ai/install.ps1 | iex

Prefer pip? pip install riva · view script

Uninstall: curl -fsSL https://rivalabs.ai/uninstall.sh | sh · Windows: irm https://rivalabs.ai/uninstall.ps1 | iex

GitHub Getting Started Riva for Enterprise

See it running

The Riva console on your machine, in the browser or the terminal, and Riva Server for the whole team.

Riva console in the browser: four running agents (Claude Code, Codex CLI, Cursor, Gemini CLI) with PID, CPU, memory, uptime and working directory, plus live CPU and memory charts per agent. — **Riva console · web** Every agent on this machine, live: status, resources, who launched it and where it's working.

Riva console in the terminal (riva watch): agent table and a detail card per running agent with CPU and memory history. — **Riva console · terminal** The same view in `riva watch`, for SSH sessions and tmux panes.

Riva Server dashboard: fleet overview with total clients, online machines, detected agents and events, and a table of connected machines with their agents. — **Riva Server · fleet** Link machines with `riva link start` and see every agent across the team in one place.

Screenshots show demo data.

13 Agent Frameworks

Drop-in observability for the agents you already run.

Claude Code Codex CLI Gemini CLI OpenClaw LangGraph CrewAI AutoGen Cursor GitHub Copilot Windsurf Continue.dev Cline OpenCode

Features

Full-stack observability for AI agents.

📊

Real-Time Monitoring

Live TUI dashboard showing agent activity, CPU/memory usage, tool calls, and execution traces as they happen.

💰

Token Analytics

Track token usage per model, cost per agent run, daily breakdowns, and top tools by call count.

🔍

Security Audit

15+ automated checks — credential exposure, permission hardening, MCP supply chain risks, prompt injection detection, network anomalies.

⏪

Session Forensics

Turn-by-turn timelines, pattern detection (dead ends, retry loops, thrashing), decision-point extraction, and cross-session efficiency trends.

⌨️

CLI / TUI

Terminal-native interface. Scan, watch, audit, inspect forensics, query history — without leaving your terminal.

🌐

Web Dashboard

Browser UI with REST API. Overview, network, security, usage, config, and forensics tabs. Dark/light themes.

🖥️

System Tray

Native macOS menu bar app. Glanceable agent status, alerts, and quick actions. Compiled Swift binary.

📡

OpenTelemetry

Export metrics, logs, and traces to any OTel-compatible backend. Bundled Grafana dashboard. All three signal types supported.

Security First

Stability and security don't come from hoping agents behave. They come from watching them.

Tamper-evident HMAC-SHA256 audit log chain
Prompt injection detection and alerting
Boundary policy engine — file, network, and process limits
Credential exposure scanning (14 token patterns)
MCP supply chain checks (HTTP endpoints, shell commands)
CEF export for SIEM integration (Splunk, QRadar)

Getting Started

# Install Riva (Windows: irm https://rivalabs.ai/install.ps1 | iex)
$ curl -fsSL https://rivalabs.ai/install.sh | sh

# Scan for running agents
$ riva scan

# Live monitoring dashboard
$ riva watch

# Run a security audit
$ riva audit

# Session forensics
$ riva forensic summary latest

# Launch the web dashboard
$ riva web start

# Link this machine to your team's Riva Server
$ riva link start https://rivalabs.your-company.ai

Running agents as a team?

Riva Server brings every machine into one command center — fleet dashboard, org-wide audit, and central policy.

Riva for Enterprise Sign in to Command Center