Security & Privacy

Security you can verify

Open source, privacy-first architecture designed for security-conscious teams. Audit our code, self-host if you want, keep your code yours.

How your data flows

Code stays local unless you explicitly share it

Your Machine

CLI daemon runs locally, watches conversation files

Code stays here
Secrets redacted
E2E encryption

Sync Layer

TLS 1.3 encrypted transport to Convex backend

TLS 1.3 in transit
Metadata only*
Paths hashed

Storage

Convex database with SOC 2 Type II compliance

SOC 2 Type II
Encrypted at rest
Your data isolated

* Full conversation content synced only if you enable it. Default: metadata only (titles, timestamps, tool names).

Security features

Built for teams that take security seriously

Private by default

Every conversation starts private. You explicitly choose what to share with teammates.

Default visibility: private

Automatic secret redaction

API keys, tokens, passwords, and credentials are automatically stripped before sync.

Pattern-based detection

End-to-end encryption

Enable client-side AES-256-GCM encryption. Your data is encrypted before it leaves your machine.

Enterprise feature

Path hashing

Project paths are hashed to prevent directory structure from being exposed.

SHA-256 hashing

Activity hiding

Make yourself invisible. Hide your activity from team dashboards and stats.

Per-user toggle

Open source

The entire codebase is open source. Audit it yourself, run security scans, verify our claims.

MIT License

What we never do

Commitments we make to every user

Train AI on your code

Your conversations and code are never used to train AI models. Period.

Sell your data

We don't sell, share, or monetize your data. Our business model is subscriptions, not data.

Access without permission

No employee can view your sessions unless you explicitly grant support access.

Store your encryption keys

When E2E encryption is enabled, keys are generated and stored only on your devices.

Technical details

For security teams and auditors

Encryption

Algorithm
AES-256-GCM
IV
Random 12-byte per encryption
Key derivation
HMAC-SHA512, BIP32-style hierarchy
Key size
256-bit (32 bytes)

Infrastructure

Transport
TLS 1.3
Backend
Convex (SOC 2 Type II)
Storage encryption
AES-256 at rest
Auth
OAuth 2.0, GitHub/Google SSO

// Example: Client-side encryption flow

const masterKey = crypto.getRandomValues(new Uint8Array(32));
const sessionKey = await deriveKey(masterKey, 'messages', [sessionId]);
const encrypted = await encryptAESGCM(message, sessionKey);
// Server only sees encrypted blob, never the plaintext
Self-hostable

Complete control

Deploy the entire Codecast stack on your own infrastructure. Your data never leaves your network. We provide deployment guides and enterprise support for self-hosted installations.

Frequently asked questions

Do you see my code?

Only if you enable code sharing. By default, we sync conversation metadata (titles, timestamps, tool names) but not the actual code content.

Will you train AI on my data?

No. Your data is never used for AI training. We use Anthropic's API for any AI features, and your data never goes to model training.

Can I delete my data?

Yes. You can delete individual conversations or your entire account. Deletion is permanent and complete.

Who can see my sessions?

Only you, by default. If you share a session with your team, only team members can see it. You can revoke access anytime.

Is it open source?

Yes. The CLI, web dashboard, and backend are all open source under the MIT license. You can audit every line of code.

Can I self-host?

Yes. Deploy the entire stack on your own infrastructure. We provide deployment guides and enterprise support.

Security contact

Found a vulnerability? We appreciate responsible disclosure and will work with you to address it quickly.

security@codecast.sh