OpenClaw (npm package 'openclaw') before 2026.8.1 fails to apply the originating requester's effective tool policy during Active Memory automatic… (CVE-2026-100537)
OpenClaw (npm package 'openclaw') versions before 2026.8.1 have a vulnerability where the effective tool policy of the originating requester is not properly enforced during Active Memory automatic recall. This allows a requester who is explicitly denied access to memory tools to still receive information derived from durable memory in the agent's response without directly invoking those memory tools. The issue is resolved in version 2026.8.1.
AI Analysis
Technical Summary
The OpenClaw npm package before version 2026.8.1 fails to enforce the originating requester's effective tool policy during Active Memory automatic recall. In deployments using Active Memory with requester-specific tool rules, deterministic and hidden recall paths can bypass explicit denials (toolsBySender) for memory tools such as memory_search and memory_get. Consequently, a requester who is admitted but denied access to memory tools can still receive information derived from durable memory in the agent's response without directly invoking any memory tool. This vulnerability is fixed in version 2026.8.1.
Potential Impact
An attacker or requester who is explicitly denied access to memory tools can still receive information derived from durable memory, potentially leading to unauthorized disclosure of information. The impact is limited to confidentiality (low severity) with no integrity or availability impact reported.
Mitigation Recommendations
Upgrade to OpenClaw version 2026.8.1 or later, where this issue is fixed. No other mitigation is indicated.
OpenClaw (npm package 'openclaw') before 2026.8.1 fails to apply the originating requester's effective tool policy during Active Memory automatic… (CVE-2026-100537)
Description
OpenClaw (npm package 'openclaw') versions before 2026.8.1 have a vulnerability where the effective tool policy of the originating requester is not properly enforced during Active Memory automatic recall. This allows a requester who is explicitly denied access to memory tools to still receive information derived from durable memory in the agent's response without directly invoking those memory tools. The issue is resolved in version 2026.8.1.
CVSS v3.1
Score 3.1low
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The OpenClaw npm package before version 2026.8.1 fails to enforce the originating requester's effective tool policy during Active Memory automatic recall. In deployments using Active Memory with requester-specific tool rules, deterministic and hidden recall paths can bypass explicit denials (toolsBySender) for memory tools such as memory_search and memory_get. Consequently, a requester who is admitted but denied access to memory tools can still receive information derived from durable memory in the agent's response without directly invoking any memory tool. This vulnerability is fixed in version 2026.8.1.
Potential Impact
An attacker or requester who is explicitly denied access to memory tools can still receive information derived from durable memory, potentially leading to unauthorized disclosure of information. The impact is limited to confidentiality (low severity) with no integrity or availability impact reported.
Mitigation Recommendations
Upgrade to OpenClaw version 2026.8.1 or later, where this issue is fixed. No other mitigation is indicated.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-368p-92qh-j8q4
- Osv Schema Version
- 1.4.0
- Aliases
- ["CVE-2026-100537"]
- Database Specific Severity
- LOW
- Cvss Version
- 3.1
Threat ID: 6ab74f2ff7a7c54106e1385c
Added to database: 09/26/2026, 04:50:55 UTC
Last enriched: 09/26/2026, 04:57:04 UTC
Last updated: 09/27/2026, 02:04:38 UTC
Views: 20
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.