gsccli follows SemVer. Only the latest minor of the current
major receives security fixes.
| Version | Supported |
|---|---|
| 1.x | ✅ |
| < 1.0 | ❌ |
Do not open a public GitHub issue.
Use one of the following private channels:
- Preferred — GitHub Security Advisories: https://github.com/nalyk/gsccli/security/advisories/new
- Email: dev.ungheni@gmail.com with subject prefix
[gsccli security].
Please include:
- A description of the vulnerability and its impact.
- Steps to reproduce (a minimal
gsccli …invocation, config layout, or code path). - The
gsccli --versionandnode --versionyou tested against. - Any suggested mitigation if you have one.
- Acknowledgement: within 7 days.
- Triage decision (accepted / not-a-vuln / duplicate): within 14 days.
- Fix or mitigation: target 30 days for high-severity, 90 days otherwise.
If you don't hear back, ping the email address again — mail can fail silently.
- Auth bypass — anything that uses a credential the user didn't intend to use.
- Token exfiltration —
gsccliwriting OAuth tokens or service-account JSON to a location an attacker can read, or printing them to stdout/stderr. - Cache poisoning — anything that lets one site's cache entry be served for another.
- Arbitrary file write outside
~/.gsccli/(or$GSCCLI_CONFIG_DIR) and the user's explicit-oargument. - Command injection through user-supplied input that
gscclishells out with. - MCP server escalation — anything that lets an MCP client invoke a write operation (the MCP surface is read-only by design).
- Google-side rate limits or quota exhaustion.
- API errors surfaced verbatim from
googleapis. - Output containing site URLs, GA property IDs, or other data that the authenticated user already has access to via the GSC web UI.
- Accidentally pasting a credential into a public issue (rotate it; that's not a product flaw).
Once a fix is released, the advisory is published with credit to the reporter (unless you prefer to remain anonymous). CVE assignment via GitHub when warranted.