Security

There are three ways to use Rams, and three security postures. The skill runs entirely in your editor and never reaches our servers. The MCP and the GitHub App send only the files being reviewed, over TLS, and discard your source the moment the review returns. We keep the review metadata: a score, an issue count, a duration. Never your code.

The Skill (/rams)

The most private option. Nothing leaves your machine:

• rams.md is a static text file with no code execution
• All processing happens in your local AI tool
• We never see your code, files, or the AI's responses
• No network requests, telemetry, or analytics during use

The MCP (API keys)

When your coding agent calls the hosted engine through the MCP:

• Calls are authenticated with an API key you generate in Settings. It is shown once, scoped to your workspace, and revocable at any time
• Only the files you or your agent pass to the review tool are sent, over TLS, to the Rams worker
• Files are analyzed in memory and discarded the moment the review returns. We store only the review metadata
• Each review is metered against your workspace's monthly quota
• File contents are never used to train models

The GitHub App

When you install the GitHub App for automated reviews on every pull request:

• Minimal permissions: read repository contents, write pull request reviews
• We only access the files changed in the pull request, never your whole repo
• Code is analyzed in memory and discarded immediately after the review
• We store only review metadata (score, issue count, duration), never your source
• You choose which repositories to enable. We never review a repo you have not opted in

How a review works

1. You open or update a pull request on an enabled repository (or your agent calls the MCP)
2. Rams fetches only the changed UI files, using a short-lived installation token or your API key
3. Those files are sent to the review engine over TLS and analyzed by Anthropic Claude
4. A scored review with inline fix suggestions comes back, posted on the PR or returned to your agent
5. Your source is discarded immediately. Only the metadata is kept

Infrastructure

• All communication encrypted with TLS
• GitHub App tokens are short-lived and scoped to a single installation
• API keys are workspace-scoped and revocable
• Row Level Security on all database tables
• Rate limiting on all API endpoints

Service providers

• Vercel: website hosting (SOC 2 Type II)
• Supabase: database (SOC 2 Type II)
• Railway: worker API (SOC 2 Type II)
• Anthropic: AI analysis (SOC 2 Type II, code not used for training)
• GitHub: app integration (SOC 2 Type II)
• Stripe: payments (PCI Level 1)
• Sentry: error tracking (SOC 2 Type II)

Maximum privacy

Local only: use the /rams skill. It runs entirely in your AI editor, with nothing sent to us.

Enterprise: ask us about self-hosted deployment for teams with strict compliance requirements.

Report a vulnerability

Found a security issue? Email rams@rams.ai. We respond within 24 hours.

HSLA0001 Inc.