This policy applies to the signnow/sn-mcp-server
repository and any packages published from it (e.g., the PyPI package, if applicable).
We provide security fixes for the latest minor release line (N) and the previous one (N-1). Older versions may receive fixes at our discretion.
Version line | Supported |
---|---|
0.1.x (latest) | ✅ |
< 0.1 | ❌ |
Please do not open public issues for security problems.
Use GitHub’s Private vulnerability reporting:
- Go to this repository → Security tab → Report a vulnerability.
- Fill out the advisory form with:
- Affected version(s) and environment
- Impact and clear reproduction steps (PoC if possible)
- Any suggested mitigation/fix
- (Optional) CVSS v3.1 vector and your severity assessment
- Your GitHub handle and preferred credit name
If you cannot use GitHub, you may email the maintainers at [add your security email/contact here]. Consider sharing encrypted details or a link to a secure channel.
We will acknowledge within 3 business days, triage within 7 business days, and keep you updated throughout remediation.
For confirmed issues we will:
- Assign a severity, create a private advisory, and work in a private fix branch/fork.
- Target timelines (guidelines, not guarantees):
- Critical: fix or mitigation target ≤ 14 days
- High: ≤ 30 days
- Medium: ≤ 90 days
- Low: best effort / next release
- Publish a security advisory with release notes once a fix is available, and credit the reporter unless you request otherwise.
Please give us reasonable time to remediate before any public disclosure.
In scope: vulnerabilities in this repository and its released artifacts (server binaries/containers/packages).
Out of scope: issues in SignNow production services, APIs, web apps, infrastructure, or third-party platforms. For those, use the official SignNow channels.
Good-faith, non-destructive research is welcome. Please do not:
- Perform denial-of-service, spam, or load testing against SignNow or third-party systems
- Exfiltrate or access data that is not yours
- Use social engineering, phishing, or physical intrusion
- Report issues that only affect third-party dependencies without showing exploitability in this project
- Disclose secrets or tokens belonging to real users
We will not initiate legal action for good-faith research that:
- Stays within scope and avoids privacy violations or service disruption
- Uses your own accounts/test data
- Reports findings privately and allows reasonable remediation time
- Never commit secrets; use environment variables or a secret manager
- In HTTP mode, run behind HTTPS (reverse proxy/ingress) and use short-lived tokens
- In production, provide a persistent RSA private key for OAuth via
OAUTH_RSA_PRIVATE_PEM
(rotate and back up securely) - Restrict
ALLOWED_REDIRECTS
to known URIs; use least-privilege SignNow credentials and rotate regularly
See the project README for setup and configuration details.