POST /api/analyze

Overview

The analyze endpoint performs a comprehensive AI-powered analysis of a GitHub repository, returning insights about code quality, architecture, security, dependencies, and improvement suggestions. Key Features:

Real-time streaming with Server-Sent Events (SSE)
Multi-layer caching for instant results
In-flight request deduplication
Branch-specific analysis
Daily tiered rate limiting

Endpoint

POST /api/analyze

Request

Headers

Content-Type

string

required

Must be application/json

Body Parameters

url

string

required

Full GitHub repository URLFormat: https://github.com/{owner}/{repo}Example: https://github.com/vercel/next.js

branch

string

Branch name to analyze. Defaults to the repository’s default branch (usually main or master)Example: canary, develop, main

Example Request

curl -X POST https://repolyze.ossium.live/api/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://github.com/vercel/next.js",
    "branch": "canary"
  }'

Response

The endpoint returns a streaming response using Server-Sent Events (SSE). Events are sent as they become available during analysis.

Response Headers

Content-Type: text/event-stream
Cache-Control: no-cache
Connection: keep-alive
X-RateLimit-Remaining: 9

Stream Event Types

The response stream contains multiple event types, each with a specific purpose:

1. Metadata Event

Sent first with repository information, file tree, and branch data.

type

string

"metadata"

data

object

Show metadata object

metadata

RepoMetadata

Repository metadata including name, description, stars, language, owner, etc.Fields: name, fullName, description, stars, forks, watchers, language, topics, defaultBranch, createdAt, updatedAt, pushedAt, size, openIssues, license, isPrivate, owner

fileTree

FileNode[]

Hierarchical file tree structureFileNode: { name, path, type, children?, size?, language?, extension? }

fileStats

object

File statisticsFields: totalFiles, totalDirectories, languages (object mapping language to file count)

branch

string

Name of the analyzed branch

availableBranches

BranchInfo[]

Array of available branchesBranchInfo: { name, commit: { sha, url }, protected, isDefault }

Example:

{
  "type": "metadata",
  "data": {
    "metadata": {
      "name": "next.js",
      "fullName": "vercel/next.js",
      "description": "The React Framework",
      "stars": 120000,
      "forks": 25000,
      "language": "TypeScript",
      "defaultBranch": "canary",
      "owner": {
        "login": "vercel",
        "avatarUrl": "https://avatars.githubusercontent.com/u/14985020"
      }
    },
    "fileStats": {
      "totalFiles": 1250,
      "totalDirectories": 180,
      "languages": {
        "TypeScript": 980,
        "JavaScript": 150,
        "CSS": 45
      }
    },
    "branch": "canary",
    "availableBranches": [
      {
        "name": "canary",
        "commit": { "sha": "abc123", "url": "..." },
        "protected": true,
        "isDefault": true
      }
    ]
  }
}

2. Scores Event

Contains calculated quality scores and detailed breakdown.

type

string

"scores"

data

object

Show scores object

overall

number

Overall repository score (0-100)

codeQuality

number

Code quality score (0-100)

documentation

number

Documentation score (0-100)

security

number

Security score (0-100)

maintainability

number

Maintainability score (0-100)

testCoverage

number

Test coverage score (0-100)

dependencies

number

Dependencies score (0-100)

breakdown

Record<string, object>

Detailed score breakdown by category with factorsStructure: { [category]: { score: number, factors: string[] } }

Example:

{
  "type": "scores",
  "data": {
    "overall": 92,
    "codeQuality": 95,
    "documentation": 90,
    "security": 88,
    "maintainability": 94,
    "testCoverage": 85,
    "dependencies": 90,
    "breakdown": {
      "codeQuality": {
        "score": 95,
        "factors": ["TypeScript strict mode", "Consistent formatting"]
      }
    }
  }
}

3. Automations Event

Suggested automation opportunities (issues, workflows).

type

string

"automations"

data

array

Array of automation suggestions

Show automation object

string

Unique identifier (e.g., "auto-1")

type

string

Type of automation: "issue", "pull-request", or "workflow"

title

string

Short title for the automation

description

string

Brief description

body

string

Full body text for issue/PR

labels

string[]

Suggested labels (e.g., ["ci", "automation"])

priority

string

Priority level: "low", "medium", or "high"

4. Refactors Event

Suggested code refactoring opportunities.

type

string

"refactors"

data

array

Array of refactoring suggestions

Show refactor object

string

Unique identifier (e.g., "ref-1")

title

string

Short title for the refactor

description

string

Brief description of the refactor

impact

string

Impact level: "low", "medium", or "high"

effort

string

Required effort: "low", "medium", or "high"

5. Content Event

Streamed AI-generated analysis content (text chunks).

type

string

"content"

data

string

Text chunk of AI-generated analysis

Example:

{"type":"content","data":"This is a well-structured"}
{"type":"content","data":" Next.js application with"}
{"type":"content","data":" comprehensive TypeScript support."}

6. Tier Event

User tier information for UI display.

type

string

"tier"

data

string

User tier: "anonymous", "free", or "pro"

Example:

{"type":"tier","data":"free"}

7. Done Event

Signals that the analysis is complete.

type

string

"done"

Example:

{"type":"done"}

8. Error Event

Sent if an error occurs during streaming.

type

string

"error"

data

string

Error message describing what went wrong

Example:

{"type":"error","data":"Repository not found"}

Error Responses

Non-streaming errors return standard JSON error responses:

400 Bad Request

{
  "error": "Invalid JSON in request body"
}

413 Payload Too Large

{
  "error": "Request body too large"
}

429 Too Many Requests (Burst)

{
  "error": "Too many requests. Please try again later."
}

Headers:

Retry-After: 60

429 Too Many Requests (Daily Limit)

{
  "error": "Daily limit reached. Sign in to get more analyses.",
  "code": "DAILY_LIMIT_REACHED",
  "limit": 1,
  "remaining": 0,
  "tier": "anonymous"
}

500 Internal Server Error

{
  "error": "Analysis failed. Please try again."
}

503 Service Unavailable

{
  "error": "Server is not properly configured."
}

Code Examples

JavaScript/TypeScript (Browser)

interface StreamEvent {
  type: 'metadata' | 'scores' | 'content' | 'automations' | 'refactors' | 'tier' | 'done' | 'error';
  data: any;
}

async function analyzeRepository(url: string, branch?: string) {
  const response = await fetch('https://repolyze.ossium.live/api/analyze', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ url, branch })
  });

  if (!response.ok) {
    const error = await response.json();
    throw new Error(error.error);
  }

  const reader = response.body!.getReader();
  const decoder = new TextDecoder();
  let buffer = '';

  while (true) {
    const { done, value } = await reader.read();
    if (done) break;

    buffer += decoder.decode(value, { stream: true });
    const lines = buffer.split('\n');
    buffer = lines.pop() || '';

    for (const line of lines) {
      if (line.startsWith('data: ')) {
        const event: StreamEvent = JSON.parse(line.slice(6));
        
        switch (event.type) {
          case 'metadata':
            console.log('Repository:', event.data.metadata.fullName);
            break;
          case 'scores':
            console.log('Overall score:', event.data.overall);
            break;
          case 'content':
            process.stdout.write(event.data);
            break;
          case 'done':
            console.log('\nAnalysis complete!');
            break;
          case 'error':
            console.error('Error:', event.data);
            break;
        }
      }
    }
  }
}

// Usage
analyzeRepository('https://github.com/vercel/next.js', 'canary');

Node.js

const https = require('https');

function analyzeRepository(url, branch) {
  const postData = JSON.stringify({ url, branch });
  
  const options = {
    hostname: 'repolyze.ossium.live',
    path: '/api/analyze',
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
      'Content-Length': Buffer.byteLength(postData)
    }
  };

  const req = https.request(options, (res) => {
    let buffer = '';
    
    res.on('data', (chunk) => {
      buffer += chunk.toString();
      const lines = buffer.split('\n');
      buffer = lines.pop() || '';
      
      for (const line of lines) {
        if (line.startsWith('data: ')) {
          const event = JSON.parse(line.slice(6));
          console.log(event.type, event.data);
        }
      }
    });
    
    res.on('end', () => {
      console.log('Stream ended');
    });
  });

  req.on('error', (error) => {
    console.error('Error:', error);
  });

  req.write(postData);
  req.end();
}

analyzeRepository('https://github.com/vercel/next.js', 'canary');

Python

import requests
import json

def analyze_repository(url: str, branch: str = None):
    endpoint = 'https://repolyze.ossium.live/api/analyze'
    payload = {'url': url}
    
    if branch:
        payload['branch'] = branch
    
    response = requests.post(
        endpoint,
        json=payload,
        stream=True,
        headers={'Content-Type': 'application/json'}
    )
    
    if response.status_code != 200:
        error = response.json()
        raise Exception(error.get('error', 'Unknown error'))
    
    for line in response.iter_lines():
        if line.startswith(b'data: '):
            event = json.loads(line[6:])
            event_type = event['type']
            event_data = event.get('data')
            
            if event_type == 'metadata':
                print(f"Repository: {event_data['metadata']['fullName']}")
            elif event_type == 'scores':
                print(f"Overall score: {event_data['overall']}")
            elif event_type == 'content':
                print(event_data, end='', flush=True)
            elif event_type == 'done':
                print("\nAnalysis complete!")
            elif event_type == 'error':
                print(f"Error: {event_data}")

# Usage
analyze_repository('https://github.com/vercel/next.js', 'canary')

Caching Behavior

The analyze endpoint implements sophisticated caching:

1. Analysis Result Cache

Full analysis results are cached in memory by {owner}/{repo}:{branch}. Subsequent requests for the same repository and branch receive instant cached responses. Cache Duration: In-memory, cleared on server restart

2. In-Flight Deduplication

If a repository is currently being analyzed, concurrent requests wait for the first analysis to complete and then receive the cached result. This prevents duplicate AI calls for the same repository.

3. GitHub Data Cache

Repository metadata, tree, and branches are cached server-side to reduce GitHub API calls.

Rate Limit Details

See the API Overview for complete rate limit information.

Health Check

You can check if the service is available:

curl https://repolyze.ossium.live/api/analyze

Response:

{
  "status": "ok",
  "timestamp": "2026-03-03T10:30:00.000Z",
  "services": {
    "openrouter": "configured",
    "github": "configured"
  }
}

For detailed type definitions, see Analysis Response Types.

Endpoints

Response Types

POST /api/analyze

Overview

Endpoint

Request

Headers

Body Parameters

Example Request

Response

Response Headers

Stream Event Types

1. Metadata Event

2. Scores Event

3. Automations Event

4. Refactors Event

5. Content Event

6. Tier Event

7. Done Event

8. Error Event

Error Responses

400 Bad Request

413 Payload Too Large

429 Too Many Requests (Burst)

429 Too Many Requests (Daily Limit)

500 Internal Server Error

503 Service Unavailable

Code Examples

JavaScript/TypeScript (Browser)

Node.js

Python

Caching Behavior

1. Analysis Result Cache

2. In-Flight Deduplication

3. GitHub Data Cache

Rate Limit Details

Health Check

Build docs developers (and LLMs) love

Endpoints

Response Types

​Overview

​Endpoint

​Request

​Headers

​Body Parameters

​Example Request

​Response

​Response Headers

​Stream Event Types

​1. Metadata Event

​2. Scores Event

​3. Automations Event

​4. Refactors Event

​5. Content Event

​6. Tier Event

​7. Done Event

​8. Error Event

​Error Responses

​400 Bad Request

​413 Payload Too Large

​429 Too Many Requests (Burst)

​429 Too Many Requests (Daily Limit)

​500 Internal Server Error

​503 Service Unavailable

​Code Examples

​JavaScript/TypeScript (Browser)

​Node.js

​Python

​Caching Behavior

​1. Analysis Result Cache

​2. In-Flight Deduplication

​3. GitHub Data Cache

​Rate Limit Details

​Health Check

Build docs developers (and LLMs) love

Overview

Endpoint

Request

Headers

Body Parameters

Example Request

Response

Response Headers

Stream Event Types

1. Metadata Event

2. Scores Event

3. Automations Event

4. Refactors Event

5. Content Event

6. Tier Event

7. Done Event

8. Error Event

Error Responses

400 Bad Request

413 Payload Too Large

429 Too Many Requests (Burst)

429 Too Many Requests (Daily Limit)

500 Internal Server Error

503 Service Unavailable

Code Examples

JavaScript/TypeScript (Browser)

Node.js

Python

Caching Behavior

1. Analysis Result Cache

2. In-Flight Deduplication

3. GitHub Data Cache

Rate Limit Details

Health Check