GHSA-9F72-QCPW-2HXC
Vulnerability from github – Published: 2026-03-03 19:08 – Updated: 2026-03-03 19:08Summary
In sandboxed runs, native prompt image auto-load did not honor tools.fs.workspaceOnly=true.
This optional hardening setting is not enabled by default. When operators enabled it, prompt text could still reference mounted out-of-workspace image paths (for example /agent/secret.png) and load those image bytes for vision-capable model input.
Affected Packages / Versions
- Package:
openclaw(npm) - Latest published npm version at triage time:
2026.2.23 - Vulnerable version range:
<= 2026.2.23 - Patched version (planned next release):
2026.2.24
Conditions Required
This issue required all of the following:
- sandbox mode enabled,
- tools.fs.workspaceOnly=true configured,
- an out-of-workspace mount path reachable from the sandbox (for example /agent),
- vision-capable model path active for native prompt image loading.
Technical Details
Native prompt image ingestion (detectAndLoadPromptImages / loadImageFromRef) resolved and read sandbox paths but did not apply the same workspace-root assertion used by file tools when tools.fs.workspaceOnly was set.
Fix Commit(s)
370d115549c0dadace0902775eea0d5094aedfdc
Verification
pnpm checkpnpm exec vitest run --config vitest.gateway.config.tspnpm test:fast
Release Process Note
patched_versions is pre-set to the planned next release (2026.2.24) so once npm release is available, this advisory only needs publish action.
OpenClaw thanks @tdjackey for reporting.
Publication Update (2026-02-25)
openclaw@2026.2.24 is published on npm and contains the fix commit(s) listed above. This advisory now marks >= 2026.2.24 as patched.
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 2026.2.23"
},
"package": {
"ecosystem": "npm",
"name": "openclaw"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2026.2.24"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-200",
"CWE-284"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-03T19:08:08Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "### Summary\nIn sandboxed runs, native prompt image auto-load did not honor `tools.fs.workspaceOnly=true`.\n\nThis optional hardening setting is **not enabled by default**. When operators enabled it, prompt text could still reference mounted out-of-workspace image paths (for example `/agent/secret.png`) and load those image bytes for vision-capable model input.\n\n### Affected Packages / Versions\n- Package: `openclaw` (npm)\n- Latest published npm version at triage time: `2026.2.23`\n- Vulnerable version range: `\u003c= 2026.2.23`\n- Patched version (planned next release): `2026.2.24`\n\n### Conditions Required\nThis issue required all of the following:\n- sandbox mode enabled,\n- `tools.fs.workspaceOnly=true` configured,\n- an out-of-workspace mount path reachable from the sandbox (for example `/agent`),\n- vision-capable model path active for native prompt image loading.\n\n### Technical Details\nNative prompt image ingestion (`detectAndLoadPromptImages` / `loadImageFromRef`) resolved and read sandbox paths but did not apply the same workspace-root assertion used by file tools when `tools.fs.workspaceOnly` was set.\n\n### Fix Commit(s)\n- `370d115549c0dadace0902775eea0d5094aedfdc`\n\n### Verification\n- `pnpm check`\n- `pnpm exec vitest run --config vitest.gateway.config.ts`\n- `pnpm test:fast`\n\n### Release Process Note\n`patched_versions` is pre-set to the planned next release (`2026.2.24`) so once npm release is available, this advisory only needs publish action.\n\nOpenClaw thanks @tdjackey for reporting.\n\n\n### Publication Update (2026-02-25)\n`openclaw@2026.2.24` is published on npm and contains the fix commit(s) listed above. This advisory now marks `\u003e= 2026.2.24` as patched.",
"id": "GHSA-9f72-qcpw-2hxc",
"modified": "2026-03-03T19:08:08Z",
"published": "2026-03-03T19:08:08Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/openclaw/openclaw/security/advisories/GHSA-9f72-qcpw-2hxc"
},
{
"type": "WEB",
"url": "https://github.com/openclaw/openclaw/commit/370d115549c0dadace0902775eea0d5094aedfdc"
},
{
"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:N/VA:N/SC:H/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "OpenClaw: Native prompt image auto-load did not honor tools.fs.workspaceOnly in sandboxed runs"
}
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.