getdesign is pre-1.0. Security fixes are applied to main and the most recent published release of each surface:
| Surface | Supported |
|---|---|
@getdesign/sdk |
Latest minor on npm |
@getdesign/cli |
Latest minor on npm |
api.getdesign.app |
Deployed HEAD of main |
getdesign.app web |
Deployed HEAD of main |
getdesign skill (skills/getdesign/) |
HEAD of main |
Please do not open a public GitHub issue for security reports.
Instead, use one of:
- GitHub private vulnerability reporting — Report a vulnerability (preferred).
- Email —
mohtashammurshid+security@gmail.com. Please include:
- A description of the issue and its impact
- Steps to reproduce
- The surface(s) affected (web, API, CLI, SDK, skill, shared packages)
- Any proof-of-concept code or URL
- Your preferred name for attribution (or "anonymous")
You will receive an acknowledgement within 72 hours. A triage decision and remediation plan will follow within 7 days for valid reports.
In-scope:
- Remote code execution, SSRF, or container escape in the hosted
api.getdesign.apppipeline or Daytona snapshot. - Any path that lets a user cause
getdesignto fetch internal / private-IP resources. - Supply-chain issues in the published
@getdesign/sdk/@getdesign/clipackages or theskills/getdesign/skill. - XSS, CSRF, auth bypasses in
getdesign.app. - Prompt-injection paths that can exfiltrate user data or keys from a hosted run.
Out of scope:
- Rate limiting / denial-of-service on the public free-tier API.
- Reports depending on physical access, social engineering, or compromised user devices.
- Self-XSS without a realistic delivery vector.
- Best-practice advice without a concrete impact (e.g. "you should add header X").
The getdesign agent skill runs inside the user's own coding agent (Claude Code, Codex, Cursor, etc.) using that agent's built-in tools. The skill itself has no network, file-system, or shell capability independent of the host agent. Security issues in the skill therefore reduce to:
- Prompts in
SKILL.md/TEMPLATE.mdthat could coerce the host agent into unsafe behavior. - Instructions that encourage the agent to fetch private / internal URLs.
- Instructions that would cause the agent to write files outside the working directory without user consent.
All three are in-scope and should be reported via the channels above.
We will not pursue legal action against researchers who:
- Act in good faith to identify and report vulnerabilities.
- Avoid privacy violations, destruction of data, and interruption of service.
- Give us a reasonable window to remediate before public disclosure (90 days is our default).
Thanks for helping keep getdesign and its users safe.