Getting Started with XOR GitHub App
Install in 2 minutes. First result in 15. One-click GitHub App install, first auto-review walkthrough, and engineering KPI triad.
Installation
One-click install from github.com/apps/xor-hardener. Select your org, choose repositories, and authorize. XOR creates an evergreen issue in each repo listing every capability and how to trigger it.
Your first interaction
Mention @xor-hardener in any PR comment with what you need. XOR reads the diff, checks for vulnerabilities, and posts a structured review. No configuration files. No YAML.
Install in 2 minutes. First result in 15.
XOR's GitHub App runs security analysis on your pull requests. Mention @xor-hardener in any PR or issue comment, tell it what you need, and it does the work. No configuration files. No YAML. One install, then prompt it.
Verification = reproduce the vulnerability, apply the patch, re-run the exploit harness, and reject if behavior deviates.
Engineering KPI triad
13 min
Median time to verified fix
From CVE to merged patch
45+ min
Reviewer minutes saved
Reclaimed per vulnerability
370
Broken patches caught
Rejected before review
Source: CVE-Agent-Bench, 128 CVEs, 13 agents evaluated. Full results →
Installation
Prerequisites: GitHub org admin access and at least one active repository.
- Go to github.com/apps/xor-hardener
- Click Install
- Select your org
- Choose "Only select repositories," then pick 1-5 repos to start
- Click Install & Authorize
What happens next
- XOR creates a tracking issue in each selected repo ("XOR Evergreen Issue") listing every capability and how to trigger it
- XOR sets up a read-only mirror of each repo for analysis
- You can mention
@xor-hardenerin any PR or issue to request work
Your first interaction
Open any pull request. Add a comment:
@xor-hardener Review this PR for security issues.
XOR reads the diff, runs your test suite against the proposed code, checks for new vulnerabilities, and posts a structured review with inline comments. The app analyzes the change in context. It's not a generic scanner. For every capability, see Platform Capabilities.
Data isolation
Your code stays in your org. XOR operates on a read-only mirror with no write access to your repositories. All analysis runs against this mirror. When you uninstall, the mirror is deleted.
Training data: Your code never becomes training data without explicit opt-in. XOR analyzes your dependencies; it doesn't learn from your proprietary code.
[NEXT STEPS]
Start using XOR
FAQ
How long does installation take?
Under 2 minutes. Go to github.com/apps/xor-hardener, click Install, select your org and repos. XOR creates a tracking issue in each repo and you can start using it immediately.
Does XOR have write access to my repositories?
No. XOR operates on a read-only mirror. All analysis runs against this mirror. When you uninstall, the mirror is deleted. Your code never becomes training data without explicit opt-in.
What is the first thing I should try?
Open any pull request and comment: @xor-hardener Review this PR for security issues. XOR reads the diff, checks for vulnerabilities, and posts a structured review with inline comments.
Automated Vulnerability Patching and PR Review
Patches CVEs automatically. Reviews every AI-generated PR with a pass/fail verification report.
Platform Capabilities
One install. Seven capabilities. Prompt-driven. CVE autopatch, PR review, CI hardening, guardrail review, audit packets, and more.
Dependabot Verification
Dependabot bumps versions. XOR verifies they're safe to merge. Reachability analysis, EPSS/KEV enrichment, and structured verdicts.
Compliance Evidence
Machine-readable evidence for every triaged vulnerability. VEX statements, verification reports, and audit trails produced automatically.
Compatibility and Prerequisites
Languages, build systems, CI platforms, and repository types supported by XOR. What you need to get started.
Command Reference
Every @xor-hardener command on one page. /review, /describe, /ask, /patch_i, /issue_spec, /issue_implement, and more.
See which agents produce fixes that work
128 CVEs. 15 agents. 1,920 evaluations. Agents learn from every run.