GHSA-G2F6-PWVX-R275
Vulnerability from github – Published: 2026-03-16 20:41 – Updated: 2026-03-16 20:41Summary
openclaw versions <= 2026.3.12 accepted unsanitized iMessage remote attachment paths when staging files over SCP, allowing shell metacharacters in the remote path operand.
Affected Packages / Versions
- Package:
openclaw(npm) - Affected versions:
<= 2026.3.12 - Fixed version:
2026.3.13
Details
The vulnerable path was the remote attachment staging flow in src/auto-reply/reply/stage-sandbox-media.ts. When ctx.MediaRemoteHost was set, OpenClaw staged the attachment by spawning /usr/bin/scp against <remoteHost>:<remotePath>. In affected releases, the remote host was normalized but the remote attachment path was not validated for shell metacharacters before being passed to the SCP remote operand. A sender-controlled iMessage attachment filename containing shell metacharacters could therefore trigger command execution on the configured remote host when remote attachment staging was enabled.
This issue is in scope under OpenClaw's trust model because it crosses an inbound content boundary into host command execution on a configured remote attachment host.
Fix
openclaw@2026.3.13 validates the SCP remote path before spawning scp. Current code calls normalizeScpRemotePath(...) and rejects paths containing shell metacharacters instead of passing them through to the remote shell.
Regression coverage exists in src/auto-reply/reply.stage-sandbox-media.scp-remote-path.test.ts (rejects remote attachment filenames with shell metacharacters before spawning scp).
Fix Commit(s)
a54bf71b4c0cbe554a84340b773df37ee8e959de
Thanks @lintsinghua for reporting.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2026.3.12"
},
"package": {
"ecosystem": "npm",
"name": "openclaw"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2026.3.13"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-78"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-16T20:41:12Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "### Summary\n`openclaw` versions `\u003c= 2026.3.12` accepted unsanitized iMessage remote attachment paths when staging files over SCP, allowing shell metacharacters in the remote path operand.\n\n### Affected Packages / Versions\n- Package: `openclaw` (`npm`)\n- Affected versions: `\u003c= 2026.3.12`\n- Fixed version: `2026.3.13`\n\n### Details\nThe vulnerable path was the remote attachment staging flow in `src/auto-reply/reply/stage-sandbox-media.ts`. When `ctx.MediaRemoteHost` was set, OpenClaw staged the attachment by spawning `/usr/bin/scp` against `\u003cremoteHost\u003e:\u003cremotePath\u003e`. In affected releases, the remote host was normalized but the remote attachment path was not validated for shell metacharacters before being passed to the SCP remote operand. A sender-controlled iMessage attachment filename containing shell metacharacters could therefore trigger command execution on the configured remote host when remote attachment staging was enabled.\n\nThis issue is in scope under OpenClaw\u0027s trust model because it crosses an inbound content boundary into host command execution on a configured remote attachment host.\n\n### Fix\n`openclaw@2026.3.13` validates the SCP remote path before spawning `scp`. Current code calls `normalizeScpRemotePath(...)` and rejects paths containing shell metacharacters instead of passing them through to the remote shell.\n\nRegression coverage exists in `src/auto-reply/reply.stage-sandbox-media.scp-remote-path.test.ts` (`rejects remote attachment filenames with shell metacharacters before spawning scp`).\n\n### Fix Commit(s)\n- `a54bf71b4c0cbe554a84340b773df37ee8e959de`\n\nThanks @lintsinghua for reporting.",
"id": "GHSA-g2f6-pwvx-r275",
"modified": "2026-03-16T20:41:12Z",
"published": "2026-03-16T20:41:12Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/openclaw/openclaw/security/advisories/GHSA-g2f6-pwvx-r275"
},
{
"type": "WEB",
"url": "https://github.com/openclaw/openclaw/commit/a54bf71b4c0cbe554a84340b773df37ee8e959de"
},
{
"type": "PACKAGE",
"url": "https://github.com/openclaw/openclaw"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:U",
"type": "CVSS_V4"
}
],
"summary": "OpneClaw accepts unsanitized iMessage attachment paths which allowed SCP remote-path command injection"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.