meraki2tf handles Meraki Dashboard API keys and produces snapshots of entire network configurations. Treat every security report accordingly.
Please report vulnerabilities privately via GitHub Security Advisories. Do not open a public issue. You can expect an acknowledgement within a week.
Include redacted reproduction detail — never real org IDs, serials, hostnames, or tokens.
The main branch is the supported version. There are no maintained release
branches yet.
- API keys are read from the environment (
MERAKI_DASHBOARD_API_KEY) or owner-only files — never from flags, and never logged. Credential-shaped backend-config keys are refused so no secret reaches argv. - Exactly two artifacts may contain secret values: the unsanitized snapshot and the Terraform state. Both are written owner-only (0600). Everything else — HCL kit, coverage manifest, runbook, alerts, logs — carries attribute names and locators, never values.
- The Meraki organization is never mutated outside the explicit
--confirmDR actions; every scheduled/pipeline invocation is read-only toward Meraki by contract. - Alert payloads and webhooks: webhook URLs must be HTTPS (tokens may be embedded in them); payloads never carry secret values.
- The sanitizer (
--sanitize) pseudonymizes identifying data for shareable/drill snapshots; a sanitized snapshot is refused where using it would be unsafe (e.g.--heal).
Anything that violates one of these bullets is a security bug — please report it, even if it looks minor.