Skip to main content

RayTrace

See what your coding agent did: traces, tool calls, cost and step-by-step evidence for Claude Code and Codex, and sandboxes (cloud or local) to run them in.

This is the RayTrace CLI from npm (@raytrace-cli/cli), packaged for pip with its own Node.js, so you don't need Node installed. The command is raytrace.

pip install raytrace-cli    # or: uv tool install raytrace-cli / pipx install raytrace-cli

Local sandbox (your project stays on this machine)

No account needed. From a Git repository:

raytrace sandbox create --local              # a copy of this repo in a gVisor sandbox in a Linux VM
raytrace connect <name> --claude             # a terminal in it, with Claude Code running (or --codex)
raytrace open                                # the local dashboard: what the agent ran, read and connected to

The first one sets up the VM, which takes several minutes. Local sandboxes need Lima; raytrace sandbox create --local offers to download it. macOS and Linux; Windows is experimental. raytrace doctor shows what is missing.

Cloud sandbox

raytrace auth login                          # sign in, or create your account, in the browser
raytrace sandbox create                      # upload this Git repository to a new sandbox
raytrace connect <sandbox-id> --claude       # a terminal in it, with Claude Code running (or --codex)

Sessions show up at https://app.raytracer.si.

In your editor (MCP)

Your agent can create, open, list, start, stop and destroy local and cloud sandboxes for you. Claude Code:

claude mcp add --scope user raytrace -- uvx --from raytrace-cli raytrace mcp

Codex (~/.codex/config.toml), Cursor and others take the same command:

[mcp_servers.raytrace]
command = "uvx"
args = ["--from", "raytrace-cli", "raytrace", "mcp"]

Then ask your agent, for example, "put this repo in a local sandbox and open it".

Commands

Command What it does
raytrace sandbox create [--local] [--name <name>] copy this Git repository into a new sandbox, in the cloud or (--local) on this machine
raytrace connect [id or name] [--claude | --codex] open a terminal in a sandbox, optionally with an agent running in it
raytrace sandbox list [--local] your sandboxes
raytrace sandbox start|stop <id or name> start or stop one; stopping keeps its files
raytrace sandbox destroy <id or name> delete one. Cloud: --keep-traces (paid plans). Local: --delete-logs or --keep-logs
raytrace open open the local dashboard
raytrace mcp the MCP server for your editor (above)
raytrace auth login|status|logout your RayTrace sign-in, for cloud sandboxes
raytrace upload logs [--sandbox <name>] send a local sandbox's recorded logs to the cloud dashboard; raytrace upload status shows your storage
raytrace share <trace-id> [--fork] a time-limited link to one recorded prompt; --list, --revoke <share-id>
raytrace report <trace-id> [--pr <number>] a pull-request comment for one run: commands and exit codes, files changed, cost. Prints it, or posts it with gh; --sandbox <rtp-id> for a cloud run, --link, --dry-run
raytrace export <trace-id> [--format json|markdown] [--output <file>] export a recorded prompt to the terminal or a file
raytrace --code <code> redeem a beta invite code
raytrace key [openrouter|typesafe] [--remove] your keys for the dashboard's AI features
raytrace doctor what this machine has and lacks

raytrace help lists everything.

Uninstall

raytrace sandbox destroy <id or name>        # any sandboxes you no longer need
raytrace uninstall --purge                   # local sandboxes: deletes the VM and ~/.raytrace
raytrace auth logout
pip uninstall raytrace-cli
claude mcp remove raytrace -s user           # if you added the MCP server to Claude Code

Wheels are built for macOS 13.5+ (Apple Silicon and Intel), Linux with glibc 2.28+ (x86-64 and ARM64) and Windows (x64 and ARM64).

Full documentation: https://github.com/Godcreatebugs/raytrace-cli

Metadata

Release files for raytrace-cli 0.12.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distributions (wheels)

Table of built distributions (wheels) for raytrace-cli 0.12.0
File Actions
raytrace_cli-0.12.0-py3-none-win_arm64.whl Python 3 none Windows ARM64 Details for raytrace_cli-0.12.0-py3-none-win_arm64.whl
raytrace_cli-0.12.0-py3-none-win_amd64.whl Python 3 none Windows x86-64 Details for raytrace_cli-0.12.0-py3-none-win_amd64.whl
raytrace_cli-0.12.0-py3-none-manylinux_2_28_x86_64.whl Python 3 none Linux glibc 2.28+ x86-64 Details for raytrace_cli-0.12.0-py3-none-manylinux_2_28_x86_64.whl
raytrace_cli-0.12.0-py3-none-manylinux_2_28_aarch64.whl Python 3 none Linux glibc 2.28+ ARM64 Details for raytrace_cli-0.12.0-py3-none-manylinux_2_28_aarch64.whl
raytrace_cli-0.12.0-py3-none-macosx_13_0_x86_64.whl Python 3 none macOS 13.0+ x86-64 Details for raytrace_cli-0.12.0-py3-none-macosx_13_0_x86_64.whl
raytrace_cli-0.12.0-py3-none-macosx_13_0_arm64.whl Python 3 none macOS 13.0+ ARM64 Details for raytrace_cli-0.12.0-py3-none-macosx_13_0_arm64.whl

Total release size: 242.8 MB

Release files / raytrace_cli-0.12.0-py3-none-win_arm64.whl

Download URL raytrace_cli-0.12.0-py3-none-win_arm64.whl
Size 32.2 MB
Tags Python 3 Windows ARM64
SHA-256 checksum
How to use checksums
a9d3929e53f3c2bb9ac9463481b91f5cadf79017ec8eeca822c3271d80e683e7
BLAKE2b-256 checksum
How to use checksums
32d8fa56e14c7ad4ca35650a68f1f71141692cdf78abd053ac523d690f1e725f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / raytrace_cli-0.12.0-py3-none-win_amd64.whl

Download URL raytrace_cli-0.12.0-py3-none-win_amd64.whl
Size 36.3 MB
Tags Python 3 Windows x86-64
SHA-256 checksum
How to use checksums
ccc2a886d709f5dbb4e3c4c3d6043bc1accfb4a4b7e9238742a722f9756fb8a0
BLAKE2b-256 checksum
How to use checksums
c0e1153d930f2141f6b0d54a4f40969a1df1bd7aad671dc6a345f0d5a5c604e6
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / raytrace_cli-0.12.0-py3-none-manylinux_2_28_x86_64.whl

Download URL raytrace_cli-0.12.0-py3-none-manylinux_2_28_x86_64.whl
Size 46.0 MB
Tags Linux glibc 2.28+ x86-64 Python 3
SHA-256 checksum
How to use checksums
bd70d4acd9be193c9a686cda5dd7fe2b73c7c76d0163f6ff294e990c4900544f
BLAKE2b-256 checksum
How to use checksums
cbe0f7d18a230bd8a0a982fcf07d51e665a2fe1485c0bae2f4582d08de644961
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / raytrace_cli-0.12.0-py3-none-manylinux_2_28_aarch64.whl

Download URL raytrace_cli-0.12.0-py3-none-manylinux_2_28_aarch64.whl
Size 45.7 MB
Tags Linux glibc 2.28+ ARM64 Python 3
SHA-256 checksum
How to use checksums
44880a2449fdf2eb69cc706d56edc2115e4b493176d8ee1e4052a19bcdf204cb
BLAKE2b-256 checksum
How to use checksums
52b5c37872488018b09d6d972bf8043c66d467204d328cabef67012f2a3a75e9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / raytrace_cli-0.12.0-py3-none-macosx_13_0_x86_64.whl

Download URL raytrace_cli-0.12.0-py3-none-macosx_13_0_x86_64.whl
Size 42.0 MB
Tags Python 3 macOS 13.0+ x86-64
SHA-256 checksum
How to use checksums
6b09395fcc93d79535285b8102d14143a08b2a2a7fd7c837a15e3c87aaf87403
BLAKE2b-256 checksum
How to use checksums
1590178f788d7fd0631790ddc5b605378900cacb23d94158704c81ffe3980989
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Release files / raytrace_cli-0.12.0-py3-none-macosx_13_0_arm64.whl

Download URL raytrace_cli-0.12.0-py3-none-macosx_13_0_arm64.whl
Size 40.7 MB
Tags Python 3 macOS 13.0+ ARM64
SHA-256 checksum
How to use checksums
379980f9805f5f68203bd5de195cb7c21461c581b36f6c50fe68f4f15a854375
BLAKE2b-256 checksum
How to use checksums
424eb5ccfa9f4266899cfb5b574e299821fc1750f3afbacdb42e573e886da9d3
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.7

Security for raytrace-cli 0.12.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.12.0 This release

6 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page