← Back to Skills Marketplace
553
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install github-cred
Description
Analyze GitHub user contribution quality. Evaluates owned repos, external contributions, output quality, and social influence. Triggers on "github cred", "an...
README (SKILL.md)
GitHub Cred
Analyze GitHub user contribution quality.
Triggers
github cred \x3Cusername>analyze \x3Cusername>'s github
Four-Tier Scoring System
Tier 0 — Owned High-Star Repos (30%)
| Metric | Score |
|---|---|
| Own ≥10k⭐ repo | +30 |
| Own ≥5k⭐ repo | +20 |
| Own ≥1k⭐ repo | +10 |
Cap: 30. Creator > Contributor
Tier 1 — External Contributions (30%)
| Metric | Score |
|---|---|
| Merged PR in ≥100k⭐ repo | +4/PR |
| Merged PR in ≥10k⭐ repo | +2/PR |
| Merged PR in ≥1k⭐ repo | +1/PR |
| Non-trivial (>20 lines) | ×1.5 |
Cap: 30
Tier 2 — Output Quality (20%)
| Metric | Score |
|---|---|
| Merge rate ≥70% | +10 |
| Merge rate ≥50% | +7 |
| Code ratio ≥80% | +5 |
| Diversity ≥10 repos | +5 |
Cap: 20
Tier 3 — Social Influence (20%)
| Metric | Score |
|---|---|
| ≥5k followers | +20 |
| ≥2k followers | +15 |
| ≥1k followers | +12 |
| ≥500 followers | +8 |
| ≥200 followers | +5 |
| ≥100 followers | +3 |
| \x3C100 followers | +1 |
Grades
| Grade | Score | Description |
|---|---|---|
| S | 95+ | Legend — top project creator + core contributor + influencer |
| A+ | 85-94 | Master — high-star project owner OR top project contributor |
| A | 70-84 | Senior — active developer with real contributions |
| B | 50-69 | Active — steady contributor |
| C | 30-49 | Beginner — some contributions |
| D | \x3C30 | Newbie |
Output Format
📊 GitHub Cred: @username
🏆 XX/100 (Grade)
▸ Tier 0 Owned Repos (XX/30)
👑 repo1 ⭐XXk
▸ Tier 1 External Contributions (XX/30)
🔨 org/repo ⭐XXk × N PRs
▸ Tier 2 Output Quality (XX/20)
📈 Merge rate XX% | Code ratio XX%
▸ Tier 3 Social Influence (XX/20)
👥 X,XXX followers
Two Paths to A+
- Creator path — Build one 10k+ stars project
- Contributor path — Deep contributions to multiple top projects (100k+ stars)
Mixed is strongest.
Usage Guidance
This skill is an instruction-only scoring rubric with no code and no requested credentials, but it's also vague about how it will retrieve GitHub data. Before installing or using it: 1) Ask the author how the agent will access GitHub (API vs scraping) and whether it will request a GITHUB_TOKEN; prefer a workflow that uses a personal token with minimal scopes if needed. 2) Understand that the rubric is subjective and that the agent's implementation may collect more GitHub data than you expect (PR diffs, commit metadata, repo lists). 3) Test the skill on non-sensitive or public accounts first. 4) If you are uncomfortable with an agent making arbitrary network requests, decline installation or require the skill to explicitly document API calls and required env vars.
Capability Analysis
Type: OpenClaw Skill
Name: github-cred
Version: 1.0.0
The skill bundle contains only metadata and markdown documentation (`_meta.json`, `SKILL.md`, `README.md`). There is no executable code present in the provided files. The markdown files describe the skill's functionality, scoring system, and usage, without any instructions that could lead to prompt injection against the agent, data exfiltration, or malicious execution. The installation instructions in `README.md` are for the user and do not constitute malicious behavior within the skill bundle itself.
Capability Assessment
Purpose & Capability
The stated purpose (analyzing GitHub contribution quality) matches the rubric and output format. However, the skill does not declare any GitHub API credential (GITHUB_TOKEN) or required tooling; while not strictly required, realistic collection of merged PR counts, merge rates, and cross-repo metrics usually needs API access or substantial scraping. The absence of declared credentials is notable but not definitive evidence of misbehavior.
Instruction Scope
SKILL.md is a scoring specification and output template but contains no concrete runtime instructions about where or how to fetch data (GitHub API vs. web scraping), what endpoints to call, or limits on what to collect. That vagueness grants the agent wide discretion to make network requests or gather extra context (potentially beyond what's needed), and it doesn't constrain reading of other data sources. The rubric references metrics (merged PRs, code ratio, merge rate) without defining computation methods or data scopes.
Install Mechanism
There is no install spec and no code files in the package (instruction-only), which is lower risk: nothing is written to disk by the bundle itself. README suggests cloning the author's GitHub repo as an install option, but the published skill contains no install script or downloaded artifacts.
Credentials
The skill requires no environment variables or credentials. That is minimal and reasonable in principle, but inconsistent with the scope of data collection it claims: GitHub API usage (to get PR merge info, commit sizes, follower counts at scale) typically needs a token to avoid rate limits. Because no credential is requested, the agent may either (a) perform unauthenticated, possibly brittle scraping, (b) repeatedly prompt for a token at runtime, or (c) attempt other network activity. No unrelated secrets are requested.
Persistence & Privilege
always is false and the skill does not request persistent system-level privileges or to modify other skills. It does not claim or appear to write system configuration or require elevated access.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install github-cred - After installation, invoke the skill by name or use
/github-cred - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release - Four-tier scoring system for GitHub contribution quality
Metadata
Frequently Asked Questions
What is GitHub Cred?
Analyze GitHub user contribution quality. Evaluates owned repos, external contributions, output quality, and social influence. Triggers on "github cred", "an... It is an AI Agent Skill for Claude Code / OpenClaw, with 553 downloads so far.
How do I install GitHub Cred?
Run "/install github-cred" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is GitHub Cred free?
Yes, GitHub Cred is completely free (open-source). You can download, install and use it at no cost.
Which platforms does GitHub Cred support?
GitHub Cred is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created GitHub Cred?
It is built and maintained by zerone0x (@zerone0x); the current version is v1.0.0.
More Skills