AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
MCP verified MIT Self-run

Traceforge

mcp-silent-night-no-trace-traceforge · by silent-night-no-trace

Open-source replay and regression testing for AI agents.

↗ Website ↗ Repository
— No reviews yet
0 installs
38 views
0.0% view→install

Install

$ agentstack add mcp-silent-night-no-trace-traceforge

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • ✓ Prompt-injection patterns
  • ✓ Secret / credential exfiltration
  • ✓ Dangerous shell & filesystem operations
  • ✓ Untrusted network calls
  • ✓ Known-malicious package signatures

What it can access

  • ✓ Network access No
  • ✓ Filesystem access No
  • ✓ Shell / process execution No
  • ✓ Environment & secrets No
  • ✓ Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/mcp-silent-night-no-trace-traceforge)

Reliability & compatibility

✓ Security review passed
0 installs to date
— no reviews yet
○ 5mo ago

Declared compatibility

Claude CodeClaude DesktopCursorWindsurf

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Traceforge? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Traceforge

[English](README.md) | [简体中文](README.zh-CN.md)

> Open-source replay and regression testing for AI agents. > > Capture real runs. Replay failures. Export tests.

Traceforge makes agent runs reproducible.

It records what actually happened across terminal, MCP, and browser tools, packages the run into a portable trace bundle, helps you inspect where it failed, and turns real failures into regression fixtures and tests.

Why

Modern agent workflows are powerful, but hard to trust.

  • terminal output is incomplete
  • MCP calls are buried in protocol logs
  • browser failures are hard to reproduce
  • expensive incidents rarely become tests
  • teams fix failures once, then see them return later

Traceforge solves one narrow problem well:

capture -> inspect -> replay -> export test

It is not another agent framework. It is the verification layer for the agent ecosystem.

What You Get

  • unified traces from terminal, MCP, and browser runs
  • a local trace viewer with timeline, artifacts, and failure boundaries
  • replay that re-runs a flow and checks where it diverges
  • export to reusable regression fixtures and test templates
  • shareable trace bundles for debugging and collaboration

CLI

traceforge capture -- node -v
traceforge view ./.traceforge/traces/run_xxxxxxxx
traceforge replay ./.traceforge/traces/run_xxxxxxxx
traceforge export-test ./.traceforge/traces/run_xxxxxxxx

Install From Source

Traceforge is currently in early alpha.

Requirements:

  • Node.js 20+
  • pnpm 10.6.0+
pnpm install
pnpm build

Development

Run the main verification commands:

pnpm typecheck
pnpm test
pnpm test:coverage
pnpm build
pnpm test:generated

Useful local entry points:

pnpm dev:cli
pnpm dev:viewer

Repository Layout

  • apps/cli - CLI entrypoint and local trace bridge
  • apps/viewer - local viewer UI
  • packages/schema - canonical trace schema and capability contract
  • packages/core - bundle, replay, and view-server helpers
  • packages/adapter-* - terminal, MCP, and browser capture adapters
  • packages/fixtures - exported regression test templates
  • examples/ - runnable examples and smoke helpers

Contributing

Contribution guidance lives in CONTRIBUTING.md.

If you change a publishable package, add a changeset with:

pnpm changeset

Security

Please report security issues privately. See SECURITY.md for the current policy.

License

MIT. See LICENSE.

Friendly Links

Repository Guide

  • Product and system design: ARCHITECTURE.md
  • Delivery plan and milestones: ROADMAP.md

Status

Early alpha.

The mission is simple:

Make any agent run reproducible.

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Write a review

Versions

  • v0.1.0 Imported from the upstream source.

More in Web & Browser