Sign in Submit site

agenttrace

github.com/luoyuctl/agenttrace AI Infrastructure & MLOps Open-source projectOpen source

agenttrace audits coding-agent sessions, tracking costs, tokens, latency, failures, and health.

Visit agenttrace 24 alternatives
github.com/luoyuctl/agenttrace
agenttrace homepage screenshot

About agenttrace

agenttrace is a local-first terminal tool for reviewing the session history of AI coding agents. It reads logs from agents such as Claude Code, Codex CLI, Qwen Code, Cline, Aider, Cursor exports, and OpenCode, along with generic JSON and JSONL traces. It provides a terminal user interface (TUI) and a command-line interface (CLI) from one binary.

Its two main uses are comparing what several agents spent on cost, tokens, and time, and diagnosing why a task ran slowly. It flags long gaps, hanging sessions, retry loops, and slow tool calls. Cost figures are estimates. Regression comparisons use a local baseline only when one is supplied. The homepage states that everything runs locally and that prompts, code, and logs do not need to leave the user's machine. Installation does download a release binary, and the install scripts verify its SHA-256 checksum.

Key features

  • Reads session logs from many coding agents, including Claude Code and Codex CLI
  • Compares sessions by agent source, model, token type, estimated cost, and wall-clock time
  • Flags long gaps, hanging sessions, retry loops, and slow tool calls
  • Ranks sessions by cost, duration, turns, health, failures, anomalies, or text search
  • Compares runs against a local baseline when one is supplied
  • Produces audit reports with raw token components, normalized pricing, and fallback confidence
Best for: Developers and teams who run several AI coding agents and want to see where session cost and time are going.
Good for
  • Runs locally and reads the logs that agents already write
  • Offers both a terminal interface and CLI output from one binary
  • Is open source
Watch out for
  • Cost figures are estimates rather than exact billing amounts
  • Regression comparison requires a local baseline to be supplied
  • Installation downloads a release binary, or requires Node.js 18+ for the npm package

agenttrace at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Open source
Yes · Source code
Language
English
Last checked
Oct 6, 2026 · Online when checked
Listed
Oct 5, 2026
open sourceai agentsai infrastructurehealth

agenttrace alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in AI Infrastructure & MLOps

Well-known listings from the same topic.

All 341
Guild AIguild.aiFlow AIflow-ai.comsystemprompt.iosystemprompt.ioCoreWeavecoreweave.comKili Technologykili-technology.comAPI docsdocs.x.ai
Related topicsAI Business ToolsAI Image GeneratorsAI Agents & AutomationAI Data & AnalyticsAI Chatbots & AssistantsAI Voice & AudioAI Writing ToolsAI Video Generators

agenttrace FAQ

Does agenttrace send my prompts, code, or logs anywhere?
The homepage states that everything runs locally and that prompts, code, and logs do not need to leave your machine. Installing the tool does involve downloading a release binary, and the install scripts verify its SHA-256 checksum.
How can agenttrace be installed?
The homepage lists install scripts for macOS, Linux, and Windows PowerShell, plus Homebrew, npm (requires Node.js 18+), and cargo install from the Git repository.
How is agenttrace updated?
Running agenttrace update checks and replaces the binary after verifying the release checksum. Package-manager installs are pointed back to their own manager. The version can be checked with agenttrace --version.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing

Listing maintained by Chris McClure and the WebIndexly editors · Updated Oct 13, 2026 · How we rank