Cloud-hosted Betterleaks v1.3.1 for hunting leaked API keys, wallet private keys, and credentials anywhere on GitHub, in cloned git repos, or in S3 / R2 / MinIO buckets. Optional live validation probes each detected secret against the vendor API to flag which ones are still active.
Available as an Apify Actor. $0.05 start + $0.05 per repo scanned + $0.30 per live-validated key. No install, no CLI, flat JSON output.
Scans a GitHub org, user, repo, pull request, issue, gist, action workflow, or S3 bucket for leaked credentials. Built on Betterleaks (the Gitleaks successor, maintained by the original Gitleaks author Zachary Rice + co-maintainers from Red Hat, Amazon, RBC). Every Betterleaks CLI flag is exposed as an input field. When you enable validation, each finding is probed live against the vendor API and tagged valid, invalid, revoked, unknown, or error.
How is it different from running Betterleaks locally?
Local Betterleaks CLI
This actor
Setup
Install Go binary, set up GitHub token, configure shell
$0.05 per repo scanned (in github / git / global_github modes)
$0.01 per finding pushed to the dataset
$0.30 per live-validated secret (only when validation is on AND status returns valid)
$0.005 per Code Search query (only in global_github mode)
Typical scans:
Scenario
Cost
Scan 1 small repo, 1 finding, no validation
$0.11
Scan 1 big repo, 30 findings, no validation
$0.40
Scan an org with 25 repos, 50 findings, no validation
$1.80
25-repo global search for rzp_live_
$1.93
100-repo scan, 200 findings, 10 live-validated
$10.05
Compare to GitGuardian Enterprise at ~$5000/year flat. Per-comprehensive-scan with this actor is 300-500x cheaper.
Which modes can I scan in?
Mode
What it scans
Required input
github
A specific GitHub org / user / repo / PR / issue / gist (you choose)
target_url, github_token recommended
global_github
All of GitHub by keyword - we add this layer via Code Search
global_search_query, github_token required
git
A single git clone URL with full history
target_url
s3
AWS S3, Cloudflare R2, MinIO, or any S3-compatible bucket
target_url, S3 creds (or s3_anonymous for public)
dir
A tarball / zip URL - we download, extract, scan filesystem
dir_source_url
stdin
Raw text content you paste in
stdin_content
Which GitHub resources can it scan in github mode?
All 12 surfaces betterleaks supports, selectable via include / exclude:
repos (default), forks
prs, pr-comments
issues, issue-comments
actions, action-artifacts
discussions
releases, release-assets
gists
You can also filter by since / until dates, exclude repos by glob pattern, or target specific GitHub Actions workflow files.
Does it find leaks that were deleted from current files?
Yes. Betterleaks scans the full git history. A secret that was committed and then deleted in a later commit still appears in the history and gets detected. The Commit field on each finding tells you which commit introduced the leak.
You can also enable advanced --log-opts pass-through to use git pickaxe (-S secret_string) for targeted historical searches.
Is the secret validated against the live vendor API?
Optional - off by default to keep scans fast. Set validation: true and Betterleaks fires HTTP requests against the vendor API for each finding using the rule's built-in CEL validator.
ValidationMeta: vendor-specific metadata. For a GitHub PAT, this includes username, name, and granted scopes. For Stripe, the test/live mode + account info.
Note: validation only fires for rules that ship with a validate CEL clause in their TOML. Not every detected secret type has one (the upstream rule set is expanding over time).
You can also tune validation behavior:
validation_timeout (per-request HTTP timeout)
validation_workers (parallel HTTP workers)
validation_status (filter output to only certain statuses)
validation_debug (include raw HTTP response in ValidationMeta)
validation_extract_empty (include validator outputs even when empty)
validation_env_vars (expose extra env vars to CEL validators)
Can I use a custom detection rule?
Yes - paste a full betterleaks TOML config into custom_config_toml. Either replaces the built-in rules or adds to them depending on how you write the TOML.
You can also use enable_rule to run only a whitelist of rule IDs (e.g. ["github-fine-grained-pat", "stripe-secret-key"]) from the built-in set without writing your own config.
Live secrets only: filter to ValidationStatus == valid, the high-signal triage list
How accurate is it?
Based on the underlying Betterleaks 1.3.1 rule set (200+ detectors), plus the upstream's BPE-tokenization based false-positive filter and CEL-based contextual filters. Compute usage observed in real test runs:
Single small repo scan: ~7 seconds, 0.0084 compute units
Q: Can I scan a private repo? Yes - provide a github_token with repo scope.
Q: Does it scan PR branches and dangling commits? PR refs yes via --include prs. Dangling commits aren't currently exposed but the underlying tool supports it.
Q: How does it handle rate limits? Without a github_token, GitHub limits to ~60 req/hr shared. With a token: 5000 req/hr on your account. Always provide a token for any real scan.
Q: Can I run it on a schedule? Yes via Apify Schedules. Recommended for continuous monitoring use cases.
Q: Can I use my own betterleaks rule TOML? Yes via custom_config_toml. Or whitelist specific built-in rules via enable_rule.
Q: What's global_github mode? Our addition on top of upstream betterleaks. You provide a search query, we use GitHub Code Search to find unique candidate repos, then run betterleaks git against each in parallel. Useful for hunting a specific secret pattern across all of GitHub.
Q: Why not use the cheaper gitleaks-github-secret-scanner actor instead? Use that one for high-volume cost-sensitive scans. Use this one when live validation matters (bug bounty triage, incident response, knowing which secrets to actually rotate).
Q: What happens to detected secrets? Written to the run's dataset, accessible only to the user who ran the actor. We never store them server-side.
Q: Can I scan an S3 bucket I don't own? Only if it's public (anonymous mode) or you have credentials.
Q: What's the maximum scan size? No hard cap on repos. max_target_megabytes (default 100) skips files larger than that.
Limitations
Per-CEL-rule validation: only rules with a validate clause actually probe live. Not every secret type has one yet.
global_github mode requires a GitHub PAT (Code Search is auth-only).
dir mode requires the source to be a downloadable tarball URL, not a local file.
Apify run timeout caps individual scans at 30 minutes per subprocess. For org scans larger than 200 repos, split into multiple runs.
The stdin mode is limited by Apify's input size cap (~5 MB).
Ethical use
For authorized security testing only. This actor scans public GitHub content (the same content any logged-in GitHub user can see), or private repos that your PAT explicitly grants access to.
Findings should be reported through responsible disclosure channels:
The leaking developer (commit author email)
The vendor's security@ address
A bug-bounty program if one exists for the vendor
Using leaked credentials without owner authorization is illegal in most jurisdictions.
Chưa có danh sách công cụ. Máy chủ từ xa sẽ có danh sách sau lần kiểm tra tình trạng đầu tiên.
Câu hỏi thường gặp
Câu hỏi
Làm thế nào để cài đặt Betterleaks Cloud GitHub & S3 Secret Scanner?
Sao chép cấu hình http cho Cursor, Claude Desktop, Claude Code hoặc VS Code trên trang này. Cấu hình dùng bản cài đặt của nhà phát hành, không phải proxy của PlugMCP.
Betterleaks Cloud GitHub & S3 Secret Scanner có cần khóa API không?
Bản ghi này không liệt kê biến môi trường bắt buộc nào.
Betterleaks Cloud GitHub & S3 Secret Scanner dùng giao thức truyền tải nào?
Betterleaks Cloud GitHub & S3 Secret Scanner được truy cập qua HTTP.
Betterleaks Cloud GitHub & S3 Secret Scanner được kiểm tra lần cuối khi nào?
Kiểm tra lần cuối: 2026-11-17. Betterleaks Cloud - GitHub Secret Scanner + Validator