Skip to the main content
The Repo Vitals mark: a pulse that ends in a commitREPO VITALS

applicable weight 65%

i-finik/codex-proxy

Language: Go
branch main

Go proxy for Codex/ChatGPT OAuth with native transport, Responses and Chat adapters, multi-account routing, and cross-platform binaries

Open on SourceCraftDownload report (Markdown)snapshot 8 October 2026signal coverage 65%analysed 22 September 2026 at 21:39 UTC — part unavailable

What the 100 is made of

category weight
Documentation 20Activity 20Security 20CI/CD 15Issues 15Code health 10

The light segments — Security: will appear with a member token; Issues: the tracker is empty or not set up. The rest share their weight for now.

Six categories

earned / weight
Documentation and best practicessignal coverage 86%
22–26.4 / 30.8
partial
Project activity
26.5 / 30.8
scored
Securitywill appear with a member token
no data
CI/CDsignal coverage 60%
13.8–23.1 / 23.1
partial
Issuesthe tracker is empty or not set up
no data
Code health
0–15.4 / 15.4
collection failed
States:
scored
partial
no data
weight goes to the rest
collection failed
interval, not zero
Not measured: CI runs
closed by the platform
will become a group once access opens
Not measured: Dockerfile, contributing guide, TODO/FIXME density
couldn't see it
the platform's window was truncated: a deep file or a recent version may have fallen outside it

Strengths

what's already done
  • SourceCraft CI is configured
  • The README introduces the project
  • Commits are fresh (16 days)
  • Commits span the whole year (101 commits)
  • Releases keep coming (16 days)

What affected the score

Penalties

Documentation and best practices

  • −3
    review policyno protection
  • −1
    branch protection policyno protection

Project activity

  • −2
    top author's share of commits — 76%
  • −1
    weekly activity — 4 weeks
  • −1
    unique authors — 3 authors

Not measured — interval width

Not a penalty: the lower bound of the score counts the metric as zero, the upper bound as full

Documentation and best practices

  • up to −3
    Dockerfiledata not collected
  • up to −1
    contributing guidedata not collected

CI/CD

  • up to −9
    CI runsdata not collected

Code health

  • up to −15
    TODO/FIXME densitydata not collected

What will raise the score

by strength of impact
  • Documentation
    +3
    set up a review policyreview policywhy: a change is reviewed, and the history stays readable
  • Activity
    +2
    share the work with other contributorstop author's share of commits — 76%why: the work does not rest on a single person
  • Documentation
    +1
    set up a branch protection policybranch protection policywhy: a change is reviewed, and the history stays readable
  • spread work across more weeks instead of bursts+1
  • bring in a second maintainer+1