ENTERPRISE PR REVIEWER UNIVERSAL AI ENGINE NODE 20+ ACTION

THE 100-EYED SENIOR ARCHITECT FOR YOUR PULL REQUESTS.

ARGUS runs an autonomous 3-stage review pipeline on every Pull Request. It maps visual control flows, verifies architectural compliance against architecture.md, and scans for technical debt before code merges into main.

argus-review-pipeline.log RUNNING

[18:30:01] INFO Starting ARGUS code review for PR #42 in owner/repo

[18:30:02] INFO Fetching PR git diff... (124 lines changed)

[18:30:03] STAGE 1 🎨 Executing Atlas (Visual Impact Topology Map)...

[18:30:04] SUCCESS Mermaid flowchart generated cleanly.

[18:30:05] STAGE 2 🏛️ Executing Athena (Architecture Compliance Guard)...

[18:30:06] COMPLIANT Verified against architecture.md parameters (0 violations).

[18:30:07] STAGE 3 ⚡ Executing Hermes (Debt & Security Scanner)...

[18:30:08] WARNING Hermes flagged 1 issue: 🟡 WARN in src/demo.js:L14 (TODO comment detected)

[18:30:09] POST Idempotent markdown review comment posted to GitHub PR #42.

BUILT FOR MODERN SOFTWARE ENGINEERING TEAMS

ARGUS is an active, production-deployed AI PR reviewer designed to eliminate code review bottlenecks, enforce architectural contracts, and guarantee 100% CI pipeline reliability.

🚀 ELIMINATE CODE REVIEW BOTTLENECK

Code reviews often slow down shipping speed. ARGUS evaluates pull requests automatically in seconds, leaving concise, single-thread executive scorecards so human reviewers focus on high-level decisions.

🛡️ ARCHITECTURAL BOUNDARY PROTECTION

As codebases grow, architectural layering degrades. ARGUS cross-references modified files against your repository's architecture.md, catching cross-layer imports and anti-patterns before code merges.

⚡ ZERO-CRASH FALLBACK ENGINE

Cloud APIs encounter rate limits or momentary network blips. ARGUS is engineered with a built-in offline static regex fallback engine that guarantees your GitHub Actions pipeline NEVER breaks or fails a build.

🌐 PRODUCT CREATION & CREDITS

Engineered and maintained by SahooShuvranshu and Crystal Studio Labs for engineering teams requiring autonomous code quality automation.

THE 3-STAGE EVALUATION PIPELINE

Every pull request is evaluated sequentially by three specialized AI personas.

STAGE 1 🎨

ATLAS — VISUALIZER

Parses raw PR git diffs and generates an interactive Mermaid.js flowchart (flowchart TD) visually mapping modified control flows and module dependency changes.

OUTPUT DIAGRAM TYPE flowchart TD
STAGE 2 🏛️

ATHENA — ARCHITECT

Cross-references code changes against rules in architecture.md. Evaluates stack declarations, layering rules, and stops architectural decay before code merges to main.

RULE ENGINE ARCHITECTURE architecture.md compliance
STAGE 3

HERMES — DEBT SCANNER

Performs line-by-line inspection to detect hardcoded API keys/secrets (🔴 BLOCK), // TODO comments and empty function stubs (🟡 WARN), and debug logs (🔵 INFO).

DETECTION MATRIX Secrets, TODOs & Placeholders

UNIVERSAL AI CLOUD PROVIDER COMPATIBILITY

Powered by Google Gemini by default, with universal support for any OpenAI-compatible provider.

PROVIDER: Google Gemini (Default)
SDK: Native @google/genai
DEFAULT MODEL: gemini-2.0-flash
BASE URL: NO (Native)
# Google Gemini (Default Workflow Setup)
- name: Run ARGUS AI Reviewer
  uses: Crystal-Studio-Labs/Argus@main
  with:
    github-token: ${{ secrets.GITHUB_TOKEN }}
    gemini-api-key: ${{ secrets.GEMINI_API_KEY }}

SELECT PRESET CODE DIFF:

INPUT: SAMPLE PR GIT DIFF
OUTPUT: ARGUS EXECUTIVE REVIEW SCORECARD

Click "RUN ARGUS EVALUATION" to execute sample code review...

DOCUMENTATION, HYBRID ENGINE & RATE LIMIT GUIDANCE

Complete setup manuals, hybrid provider auto-detection rules, action inputs, and API plan guidance.

🧠 HYBRID PROVIDER RESOLUTION & AUTO-DETECTION

ARGUS uses a smart Hybrid Resolution Engine. Explicit workflow inputs (model, base-url) ALWAYS override defaults. If omitted, ARGUS auto-routes based on key prefix:

Key Prefix Target Provider Default Base URL Default Model
gsk_... Groq Cloud https://api.groq.com/openai/v1 llama-3.3-70b-versatile
nvapi-... NVIDIA NIM https://integrate.api.nvidia.com/v1 meta/llama-3.3-70b-instruct
sk-or-... OpenRouter https://openrouter.ai/api/v1 google/gemini-2.0-flash-001
AIza... / AQ... Google Gemini Native SDK gemini-2.0-flash
⚙️ ACTION INPUTS & CONFIGURATION PARAMETERS

Reference table for all inputs declared in action.yml:

Input Key Description Required Default Value
github-token GitHub token for fetching diffs and posting review comments False ${{ github.token }}
gemini-api-key Google Gemini API key (primary default engine) False None
groq-api-key Groq Cloud API key (sub-second 70B inference) False None
api-key Universal API key for Gemini or OpenAI-compatible providers False None
base-url Custom REST endpoint URL for OpenAI-compatible providers False Auto-detected
model AI model identifier string False gemini-2.0-flash
💡 AI PROVIDER RATE LIMITS & PLAN GUIDANCE

Free Tier API Keys: Free tier API keys (such as Google Gemini Free Tier with 15 Requests/Minute) may encounter rate limit pauses (HTTP 429) during rapid consecutive PR commits.

Paid / Pro / Subscription Plans: Paid API keys or higher-tier subscription plans (Groq Cloud, NVIDIA NIM, OpenAI Pro, Gemini Pay-As-You-Go) operate with virtually unlimited rate limits and zero error pauses.

Zero CI Break Guarantee: If rate limit errors occur on Free Tier keys, ARGUS's built-in Fallback Static Analysis Engine automatically completes the PR evaluation using offline static regex parsing without crashing your CI build pipeline.

QUICK START WORKFLOW SETUP

Copy and paste this workflow file into your repository under .github/workflows/argus.yml.

.github/workflows/argus.yml
name: ARGUS AI Code Review

on:
  pull_request:
    types: [opened, synchronize, reopened]

jobs:
  argus-review:
    name: ARGUS 3-Stage AI PR Review
    runs-on: ubuntu-latest
    permissions:
      contents: read
      pull-requests: write
      issues: write

    steps:
      - name: Checkout Code
        uses: actions/checkout@v4

      - name: Setup Node.js
        uses: actions/setup-node@v4
        with:
          node-version: 20
          cache: 'npm'

      - name: Install Dependencies
        run: npm ci

      - name: Run ARGUS AI Reviewer
        uses: Crystal-Studio-Labs/Argus@main
        with:
          github-token: ${{ secrets.GITHUB_TOKEN }}
          gemini-api-key: ${{ secrets.GEMINI_API_KEY }}

MYTHOLOGICAL ORIGINS & CODEBASE MAPPING

Deeply rooted in ancient Greek mythology, engineered for modern agentic software development.

👁️

ARGUS PANOPTES

Ἄργος Πανόπτης

The 100-eyed giant who never slept. In this codebase, ARGUS (index.js) watches over every Pull Request event to prevent bugs, architectural decay, and unfinished code from reaching production.

Code Mapping: index.js (Action Orchestrator)
🎨

ATLAS

Ἄτλας

The Titan who holds up the heavens. In this codebase, Atlas (skills/atlas.js) maps the structural topology of your codebase using interactive Mermaid.js flowcharts.

Code Mapping: skills/atlas.js (Stage 1 Visualizer)
🏛️

ATHENA

Ἀθηνᾶ

The goddess of wisdom and strategic order. In this codebase, Athena (skills/athena.js) acts as the strategic Architecture Guardian enforcing strict compliance against architecture.md.

Code Mapping: skills/athena.js (Stage 2 Compliance)

HERMES

Ἑρμῆς

The swift messenger god who inspects hidden corners. In this codebase, Hermes (skills/hermes.js) inspects modified files line-by-line for TODOs, secrets, and empty function stubs.

Code Mapping: skills/hermes.js (Stage 3 Debt Scanner)