Skill

GitGauge

Score GitHub repos for authenticity. Mirrors the GitGauge Chrome extension algorithm exactly.

The GitGauge output line is the announcement — no separate announce needed. Run the scorer first, then engage with the repo.

When to run

Trigger	Output mode
GitHub URL appears passively in conversation	Compact one-liner, then continue
User asks to review / analyze a repo	Compact score + brief breakdown, then review
User runs /gitgauge owner/repo	Full report: sub-scores, flags, metadata

Do not run on:

• GitHub profile URLs, org pages, gist links, issue links, PR links, or commit links
• npm package references (@scope/package), file paths, or bare owner/repo with no GitHub URL context
• Repos already scored in the current conversation

Running the scorer

python ~/.claude/skills/gitgauge/scripts/score.py owner/repo
# macOS / Linux: use python3 instead of python

With a token (removes rate limit):

python ~/.claude/skills/gitgauge/scripts/score.py owner/repo --token ghp_yourtoken

For the full report mode (parse JSON to build structured output):

python ~/.claude/skills/gitgauge/scripts/score.py owner/repo --json

Pass full GitHub URLs too — the script strips the github.com/ prefix automatically.

Output formats

Compact (passive trigger)

GitGauge: 🟢 4.1/5.0 — Likely Authentic

Standard (review / analyze trigger)

GitGauge: 🟢 4.1/5.0 — Likely Authentic
⭐ 2,400 stars  🍴 310 forks  👁 47 watchers  🐛 18 issues
Fork ratio: 12.9%  |  Watcher ratio: 1.96%  |  Issue ratio: 0.75%

Full report (/gitgauge command)

Use --json output. Show score, label, all ratios, sub-scores, flags, and metadata (language, description, topics).

Score labels

Score	Label	Emoji
1.0–1.4	Likely Fake	🔴
1.5–2.4	Suspicious	🟠
2.5–3.4	Mixed Signals	🟡
3.5–4.4	Likely Authentic	🟢
4.5–5.0	Highly Authentic	✅

Warning threshold

If score < 2.5, show a soft warning before engaging with the repo:

⚠️ GitGauge flagged owner/repo as Suspicious (1.8/5.0) — low engagement relative to star count may indicate star farming or an inactive project. Proceeding anyway.

Never block the user — the score is additive context, not a gate.

Error handling

Error	What to show
404 / not found	GitGauge: ❌ Repo not found or private — skipping score
Rate limited	GitGauge: ⚠️ Rate limited — pass --token to continue (see Token Setup below)
Unscoreable (< 50 stars)	GitGauge: ⚪ Unscoreable — fewer than 50 stars
Bad token	GitGauge: ❌ GitHub token invalid or expired

In all error cases: continue engaging with the repo normally.

Edge case flags

The script returns these flags in output — always surface them:

Flag	What to tell the user
is_fork	"This is a fork — metrics reflect fork activity, not the original repo"
archived	"Repo is archived — activity metrics are frozen"
low_stars	"Under 100 stars — signal may not be reliable"
new_repo	"Under 30 days old — insufficient history for a reliable score"

Zero issues + many stars: Script automatically scores this as 1.5 (Suspicious). Large repos with zero open issues usually have issues disabled, not zero bugs.

Token setup (optional)

Without a token: 60 GitHub API requests/hour per IP. Sufficient for most sessions.

1. Create a token at https://github.com/settings/tokens — no scopes needed for public repos
2. Pass inline: --token ghp_yourtoken
3. Or set GITHUB_TOKEN in your environment and update the Bash call to read $GITHUB_TOKEN

Scoring algorithm

Full implementation is in scripts/score.py — a direct port of the GitGauge Chrome extension.

• Minimum threshold: repos with < 50 stars return unscoreable
• Fork score (weight: 35% with watchers, 55% without): forks ÷ stars
• Watcher score (weight: 35%, omitted if watchers = 0): watchers ÷ stars
• Issue score (weight: 30% with watchers, 45% without): issues ÷ stars
• Final score clamped to [1.0, 5.0], rounded to one decimal

Breakpoints and interpolation logic are defined as constants at the top of score.py.

Quick reference

Situation	What to do
GitHub URL in message	Run scorer → compact one-liner → continue
"Review this repo"	Run scorer → standard output → review
/gitgauge owner/repo	Run scorer with --json → full report
Score < 2.5	Show warning before engaging, then proceed
Any API error	Show error line, still engage with repo
Flag in output	Always mention it to the user
Already scored this session	Skip — don't score twice