home / skills / yoanbernabeu / supabase-pentest-skills / supabase-detect

supabase-detect skill

/skills/detection/supabase-detect

This skill detects Supabase usage in a web app by analyzing domain patterns, client libraries, and API endpoints to determine project presence.

npx playbooks add skill yoanbernabeu/supabase-pentest-skills --skill supabase-detect

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
9.9 KB
---
name: supabase-detect
description: Detect if a web application uses Supabase by analyzing client-side code, network patterns, and API endpoints.
---

# Supabase Detection

> šŸ”“ **CRITICAL: PROGRESSIVE FILE UPDATES REQUIRED**
>
> You MUST write to context files **AS YOU GO**, not just at the end.
> - Write to `.sb-pentest-context.json` **IMMEDIATELY after each discovery**
> - Log to `.sb-pentest-audit.log` **BEFORE and AFTER each action**
> - **DO NOT** wait until the skill completes to update files
> - If the skill crashes or is interrupted, all prior findings must already be saved
>
> **This is not optional. Failure to write progressively is a critical error.**

This skill determines whether a web application uses Supabase as its backend.

## When to Use This Skill

- Starting a security audit on an unknown application
- Verifying Supabase usage before running other audit skills
- Quickly checking multiple applications for Supabase presence

## Prerequisites

- Target URL must be publicly accessible
- Internet connection to fetch and analyze the target

## Detection Methods

The skill uses multiple detection vectors:

### 1. Domain Pattern Matching

Searches for Supabase-related domains in:
- HTML source code
- JavaScript bundles
- Network requests (via inline scripts)

**Patterns detected:**
```
*.supabase.co
*.supabase.com
supabase-cdn.com
```

### 2. JavaScript Client Detection

Looks for Supabase client library signatures:

```javascript
// Import patterns
import { createClient } from '@supabase/supabase-js'
const { createClient } = require('@supabase/supabase-js')

// Client initialization
supabase.createClient(
createClient('https://
SUPABASE_URL
NEXT_PUBLIC_SUPABASE
VITE_SUPABASE
REACT_APP_SUPABASE
```

### 3. API Endpoint Detection

Checks for characteristic Supabase endpoints:

```
/rest/v1/
/auth/v1/
/storage/v1/
/realtime/v1/
/functions/v1/
```

### 4. Response Header Analysis

Looks for Supabase-specific headers:

```
x-supabase-*
sb-*
```

## Usage

### Basic Detection

```
Check if https://myapp.example.com uses Supabase
```

### Detection with Verbose Output

```
Detect Supabase on https://myapp.example.com with full details
```

## Output Format

### Supabase Detected

```
═══════════════════════════════════════════════════════════
 SUPABASE DETECTED
═══════════════════════════════════════════════════════════

 Target: https://myapp.example.com
 Status: āœ… Supabase usage confirmed

 Detection Evidence:
 ā”œā”€ā”€ Domain: abc123def.supabase.co (found in main.js)
 ā”œā”€ā”€ Client: @supabase/supabase-js v2.x detected
 ā”œā”€ā”€ Endpoints: /rest/v1/, /auth/v1/, /storage/v1/
 └── Headers: x-supabase-api-version present

 Project Reference: abc123def
 Project URL: https://abc123def.supabase.co

 Context saved to: .sb-pentest-context.json
═══════════════════════════════════════════════════════════
```

### Supabase Not Detected

```
═══════════════════════════════════════════════════════════
 DETECTION RESULT
═══════════════════════════════════════════════════════════

 Target: https://myapp.example.com
 Status: āŒ Supabase not detected

 Scanned:
 ā”œā”€ā”€ HTML source: No Supabase patterns
 ā”œā”€ā”€ JavaScript bundles: 3 files analyzed, no matches
 ā”œā”€ā”€ Network patterns: No Supabase endpoints
 └── Response headers: No Supabase headers

 Note: The app may use a self-hosted Supabase or custom domain.
       Try providing a known Supabase URL manually if you have one.
═══════════════════════════════════════════════════════════
```

## Context Output

When Supabase is detected, the skill saves to `.sb-pentest-context.json`:

```json
{
  "target_url": "https://myapp.example.com",
  "detection": {
    "detected": true,
    "confidence": "high",
    "timestamp": "2025-01-31T10:00:00Z",
    "evidence": [
      {
        "type": "domain",
        "value": "abc123def.supabase.co",
        "location": "/static/js/main.js",
        "line": 1247
      },
      {
        "type": "client_library",
        "value": "@supabase/supabase-js",
        "version": "2.x"
      }
    ]
  },
  "supabase": {
    "project_ref": "abc123def",
    "project_url": "https://abc123def.supabase.co"
  }
}
```

## Audit Log Entry

Each detection is logged to `.sb-pentest-audit.log`:

```
[2025-01-31T10:00:00Z] DETECTION_START target=https://myapp.example.com
[2025-01-31T10:00:01Z] FETCH_HTML status=200 size=45KB
[2025-01-31T10:00:02Z] FETCH_JS file=main.js status=200 size=1.2MB
[2025-01-31T10:00:03Z] PATTERN_MATCH type=domain value=abc123def.supabase.co
[2025-01-31T10:00:03Z] DETECTION_COMPLETE result=detected confidence=high
```

## Confidence Levels

| Level | Criteria |
|-------|----------|
| **High** | Multiple evidence types (domain + client + endpoints) |
| **Medium** | Single strong evidence (domain or explicit client init) |
| **Low** | Only indirect evidence (generic patterns, possible false positive) |

## Edge Cases

### Custom Domains

Some Supabase projects use custom domains (e.g., `api.mycompany.com`). In this case:

```
Detect Supabase on https://myapp.com with custom API domain api.mycompany.com
```

### Self-Hosted Supabase

Self-hosted instances won't have `.supabase.co` domains. Look for:
- PostgREST patterns (`/rest/v1/`)
- GoTrue auth patterns (`/auth/v1/`)
- Supabase client library in code

### Single Page Applications

For SPAs with lazy-loaded chunks:

```
Detect Supabase on https://myapp.com including all JS chunks
```

## Common Issues

āŒ **Problem:** Detection returns false negative on SPA
āœ… **Solution:** The app may lazy-load Supabase. Try interacting with the app first to load all chunks, or provide known patterns.

āŒ **Problem:** Multiple Supabase projects detected
āœ… **Solution:** This can happen with multi-tenant setups. The skill will list all found projects.

āŒ **Problem:** Detection is slow
āœ… **Solution:** Large JS bundles take time to analyze. Use `--quick` mode for faster but less thorough detection:
```
Quick detect Supabase on https://myapp.com
```

## Next Steps

After detection:
1. Run `supabase-extract-url` to confirm and extract the project URL
2. Run `supabase-extract-anon-key` to find the API key
3. Or use `supabase-pentest` for a full guided audit

## MANDATORY: Progressive Context File Updates

āš ļø **This skill MUST update tracking files PROGRESSIVELY during execution, NOT just at the end.**

### Critical Rule: Write As You Go

**DO NOT** batch all writes at the end. Instead:

1. **Before starting any action** → Log the action to `.sb-pentest-audit.log`
2. **After each discovery** → Immediately update `.sb-pentest-context.json`
3. **After each significant step** → Log completion to `.sb-pentest-audit.log`

This ensures that if the skill is interrupted, crashes, or times out, all findings up to that point are preserved.

### Required Actions (Progressive)

1. **Create/Update `.sb-pentest-context.json`** with results:
   ```json
   {
     "target_url": "https://myapp.example.com",
     "detection": {
       "detected": true,
       "confidence": "high",
       "timestamp": "...",
       "evidence": [ ... ]
     },
     "supabase": {
       "project_ref": "abc123def",
       "project_url": "https://abc123def.supabase.co"
     }
   }
   ```

2. **Create/Log to `.sb-pentest-audit.log`**:
   ```
   [TIMESTAMP] [supabase-detect] [START] Starting Supabase detection
   [TIMESTAMP] [supabase-detect] [SUCCESS] Supabase detected with high confidence
   [TIMESTAMP] [supabase-detect] [CONTEXT_UPDATED] .sb-pentest-context.json created/updated
   ```

3. **IMPORTANT: As the first skill in the audit chain, this skill is responsible for creating the context files if they don't exist.**

**FAILURE TO UPDATE CONTEXT FILES IS NOT ACCEPTABLE.**

## MANDATORY: Evidence Collection

šŸ“ **Evidence Directory:** `.sb-pentest-evidence/01-detection/`

### Evidence Files to Create

| File | Content |
|------|---------|
| `initial-scan.json` | Raw detection results with all evidence |
| `supabase-endpoints.txt` | List of discovered Supabase endpoints |
| `client-code-snippets/` | Directory with relevant code excerpts |

### Evidence Format

```json
{
  "evidence_id": "DET-001",
  "timestamp": "2025-01-31T10:00:00Z",
  "category": "detection",
  "target_url": "https://myapp.example.com",

  "detection_results": {
    "supabase_detected": true,
    "confidence": "high",
    "project_url": "https://abc123def.supabase.co",
    "project_ref": "abc123def"
  },

  "evidence": [
    {
      "type": "domain_pattern",
      "value": "abc123def.supabase.co",
      "location": "/static/js/main.js",
      "line": 1247,
      "context": "const SUPABASE_URL = 'https://abc123def.supabase.co'"
    },
    {
      "type": "client_library",
      "value": "@supabase/supabase-js",
      "version": "2.x"
    }
  ],

  "curl_command": "curl -s 'https://abc123def.supabase.co/rest/v1/' -H 'apikey: [ANON_KEY]'"
}
```

### Add to curl-commands.sh

```bash
# === DETECTION ===
# Check Supabase API availability
curl -s "$SUPABASE_URL/rest/v1/" -H "apikey: $ANON_KEY" | head -100
```

### Add to timeline.md

```markdown
## [TIMESTAMP] - Detection Phase Complete
- Supabase detected with [confidence] confidence
- Project: [project_ref]
- Evidence: `01-detection/initial-scan.json`
```

## Related Skills

- `supabase-extract-url` — Extract project URL from code
- `supabase-extract-anon-key` — Find anon key
- `supabase-pentest` — Full orchestrated audit

Overview

This skill detects whether a web application uses Supabase by analyzing client-side code, network patterns, and API endpoints. It performs automated checks across HTML, JavaScript bundles, network responses, and headers, then produces structured evidence and a confidence rating. The skill is designed as the first step in a Supabase-focused security audit and creates progressive context and audit logs as it runs.

How this skill works

The detector scans the target URL's HTML and JavaScript files for Supabase domain patterns and client library signatures (e.g., @supabase/supabase-js and createClient initialization). It probes characteristic API paths (/rest/v1/, /auth/v1/, /storage/v1/, /realtime/v1/, /functions/v1/) and inspects response headers for Supabase-specific identifiers. Findings are saved progressively to context and audit log files and bundled into an evidence directory for follow-up actions.

When to use it

  • At the start of a security audit to confirm if Supabase is used
  • Before running Supabase-specific audit steps like key extraction or RLS testing
  • When triaging multiple targets to prioritize Supabase-related assessments
  • To verify Supabase usage in single-page applications or lazy-loaded sites
  • When investigating possible self-hosted or custom-domain Supabase setups

Best practices

  • Run an interactive crawl or exercise the SPA to load lazy-loaded JS chunks before scanning
  • Enable verbose mode to capture client initialization snippets and full endpoint lists
  • Save findings progressively: write context and audit log entries immediately after each discovery
  • Collect code excerpts and endpoint lists into an evidence directory for reproducibility
  • Cross-check domain matches with observed endpoints to reduce false positives

Example use cases

  • Quickly determine whether myapp.example.com relies on Supabase before deeper testing
  • Scan a batch of customer URLs and flag those that expose Supabase endpoints
  • Confirm a suspected Supabase project_ref found in obfuscated JS bundles
  • Detect self-hosted Supabase by identifying PostgREST or GoTrue endpoint patterns
  • Produce an evidence package (initial-scan.json, endpoints list, code snippets) for audit records

FAQ

What confidence levels does the skill report and what do they mean?

High: multiple evidence types (domain + client + endpoints). Medium: a single strong indicator (explicit client init or project domain). Low: indirect or ambiguous patterns that may be false positives.

How does the skill handle single-page apps and lazy-loaded chunks?

The skill supports interactive crawling; exercising the app can load lazy bundles. If chunks still miss detection, provide known Supabase patterns or run a deeper JS bundle analysis.

What output and artifacts are produced?

Progressive context (.sb-pentest-context.json), audit log entries (.sb-pentest-audit.log), and an evidence directory with initial-scan.json, supabase-endpoints.txt, and client code snippets for reproducibility.