GHSA-chm2-m3w2-wcxm
OpenClaw Google Chat spoofing access with allowlist authorized mutable email principal despite sender-ID mismatch
Auswirkung
### Summary Google Chat allowlisting supports matching by sender email in addition to immutable sender resource name (`users/<id>`). This weakens identity binding if a deployment assumes allowlists are strictly keyed by immutable principals. ### Affected Packages / Versions (As of 2026-02-14; based on latest published npm versions) - `openclaw` (npm): `<= 2026.2.13` - `clawdbot` (npm): `<= 2026.1.24-3` ### Details Affected component: - `extensions/googlechat/src/monitor.ts` The `allowFrom` checks accept: - Immutable sender id (`users/<id>`) - Raw email (`alice@example.com`) for usability Historically, `users/<email>` was also treated as an email allowlist entry. This is now deprecated because it looks like an immutable ID but is actually a mutable principal. ### Security Triage (2026-02-14) Severity: **Low** Rationale: - Requests are authenticated as coming from Google Chat (token verification), so this is not a generic unauthenticated spoofing vector. - A realistic exploit generally requires **Google Workspace / IdP administrative control** over identity lifecycle (e.g. reassigning an email address to a different underlying account) to obtain the same email with a different `users/<id>`. - With that level of access, the attacker typically has broader compromise paths. We still treat it as a valid defense-in-depth report because accepting mutable principals in authorization decisions can increase risk in chained-failure scenarios. ### Remediation / Behavior Changes Goal: preserve usability while reducing footguns. - Raw email allowlists remain supported. - `users/<email>` is deprecated and treated as a **user id**, not as an email allowlist. - Documentation recommends `users/<id>` when strict immutable binding is required. ### Fix Commit(s) - `c8424bf29a921e25663b29f308640b3d91a49432` (PR #16243) Thanks @vincentkoc for reporting.
Ein Angreifer kann die Lücke über das Netzwerk ausnutzen, mit einem Administratorkonto, ohne Zutun eines Nutzers. Dafür müssen bestimmte Bedingungen erfüllt sein, was einen Angriff erschwert. Bei Erfolg: geringes Risiko, dass Daten ausgelesen werden, geringes Risiko, dass Daten oder Systeme verändert werden.
Betroffene Pakete
| Ökosystem | Paket | Betroffen | Behoben in |
|---|---|---|---|
| npm | openclaw | < 2026.2.14 | 2026.2.14 |
| npm | clawdbot | <= 2026.1.24-3 | nicht erfasst |
Schwachstellenklasse
Referenzen
- GitHub-Advisory
- https://github.com/advisories/GHSA-chm2-m3w2-wcxm
- https://github.com/openclaw/openclaw/security/advisories/GHSA-chm2-m3w2-wcxm
- https://github.com/openclaw/openclaw/pull/16243
- https://github.com/openclaw/openclaw/commit/c8424bf29a921e25663b29f308640b3d91a49432
- https://github.com/openclaw/openclaw/releases/tag/v2026.2.14