npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

agent-run-trace-pack

v0.1.1

Published

Local-first CLI for redacted, reviewable coding-agent trace packs.

Downloads

207

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Git

Maintainers

aolinggeaolingge

Keywords

ai-agentcoding-agentcoding-agentsagentic-codingobservabilitytraceaudit-logmcpdeveloper-toolscligitredactionlocal-firsttypescript

Readme

Agent Run Trace Pack

Agent Run Trace Pack

Record local coding-agent runs into redacted, reviewable trace packs.

Status CI License: MIT Node

Agent Run Trace Pack is a local-first CLI for maintainers and teams using Codex, Claude Code, Cursor, Gemini CLI, OpenCode, Goose, GitHub Copilot coding workflows, or MCP-based tooling. It wraps a command, records what happened, redacts sensitive output, captures git state, and writes a shareable report.

Quick Start

npx agent-run-trace-pack run -- npm test

Run from source when contributing:

npm install
npm run build
node dist/cli.js run -- npm test

Docs

Package Contract

Current package metadata targets Node >=20 and exposes agent-run-trace-pack and artrace.

The package preview is limited to the npm files allowlist: dist, docs, assets, README.md, and LICENSE, plus npm's required package metadata. docs/release-readiness.md records the exact local checks used to verify that preview.

The trace pack writes:

  • .agent-traces/<trace-id>/manifest.json
  • .agent-traces/<trace-id>/stdout.log
  • .agent-traces/<trace-id>/stderr.log
  • .agent-traces/<trace-id>/diff.patch
  • .agent-traces/<trace-id>/report.md
  • .agent-traces/<trace-id>/report.html

The JSON report is manifest.json; no separate report.json file is written in v0.1.

Why It Exists

Coding agents increasingly read files, run commands, call tools, and edit repositories. When a run succeeds, fails, leaks a token-like value, or changes more than expected, maintainers need a compact local record that explains the command, output, diff, exit status, and risk signals without uploading source code.

CLI

agent-run-trace-pack run [--out .agent-traces] [--cwd path] -- <command...>
agent-run-trace-pack summarize <trace-dir>
agent-run-trace-pack doctor
agent-run-trace-pack init

Examples:

artrace run -- npm test
artrace run --out .tmp/traces -- codex --help
artrace run -- gemini --help
artrace summarize .agent-traces/2026-04-28T10-00-00-000Z

More examples are in docs/examples.md, including synthetic Codex, Claude Code, Gemini CLI, OpenCode, and Goose-style commands.

What It Captures

| Surface | What gets recorded | | --- | --- | | Command | Wrapped command, working directory, start/end time, duration, exit code | | Output | Redacted stdout and stderr logs | | Git state | HEAD, branch, status, diff stat, and patch after the run | | Risk signals | Destructive commands, pipe-to-shell, publish/release commands, permission failures, secret-like output | | Reports | JSON manifest, Markdown report, and HTML report |

Safety Model

  • Local-first: no telemetry, hosted API, model call, or cloud upload.
  • Redacted output: token-like values are removed from saved logs and reports.
  • Review before sharing: trace packs can still reveal file names, command names, and project structure.
  • No replay by default: the tool records a run; it does not automatically rerun or approve commands.

Report Preview

Report preview

Development

npm install
npm run check
npm run smoke
node scripts/release-dry-run.mjs

Repository layout:

src/
  cli.ts
  core/
  report/
tests/
docs/
assets/

Roadmap

  • v0.1: CLI trace packs, redaction, git diff capture, Markdown/HTML/JSON reports.
  • v0.2: GitHub Action wrapper and PR artifact upload examples.
  • v0.3: MCP gateway policy recording and tool-call schema snapshots.
  • v0.4: multi-agent eval runs across Codex, Gemini CLI, OpenCode, Claude Code, and Goose.

Security

Do not share trace packs that contain private source details, private URLs, raw logs, cookies, browser profiles, or real credentials. Redaction is a safety net, not a guarantee.

Contributing

Small, well-tested contributions are welcome. Start with CONTRIBUTING.md, run npm run check, and include a synthetic trace fixture when changing detection behavior.

License

MIT