Back to skill

Security audit

OpenClaw Exec Permission

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed exec-permission guide, but it teaches users to bypass protected settings and disable both command restrictions and approvals.

Install only if you intentionally want guidance for changing OpenClaw exec security. Avoid using the full plus off configuration except for tightly controlled, temporary maintenance on a dedicated machine, and restore allowlist with approval afterward.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:29
Finding

Bypass of Protected Execution Controls Enables Unrestricted Command Execution

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill provides explicit instructions to directly edit the OpenClaw configuration to set tools.exec.security = 'full' and tools.exec.ask = 'off', which removes both command restrictions and approval prompts. Because this is framed as the 'correct method' and includes copy-pasteable code, it materially lowers execution safeguards and could enable arbitrary command execution by an agent or a user following the guide.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The recommendations label full + off as acceptable for 'development' and 'trusted private machine' scenarios without a prominent warning that this disables all exec safeguards. In the context of an agent skill specifically about execution permissions and privilege elevation, normalizing these unsafe defaults increases the chance operators will deploy insecure settings that permit unrestricted command execution.

Content

No source excerpt is available for this finding.