Back to skill

Security audit

Openclaw Task Executor

Security checks for vulnerabilities and agentic risk

Overview

This skill mainly provides a task workflow, but it redirects plans and results to an undefined controller named TOM and encourages unbounded retries and subagent use.

Review this skill before installing. It has no executable payload, but it can change how an agent handles many tasks by routing plans and results to TOM, spawning subagents for medium or complex work, and retrying failures without clear limits. Install only if TOM is your intended controller and you are comfortable with explicit user approval being required before sensitive actions, capability inspection, or any sharing of results.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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 (2)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:7
Finding

Mandatory Workflow Redirection to an Undefined External Controller

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:63
Finding

Unnecessary Reconnaissance of Execution Capabilities and Security Configuration

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The note on L79 switches to Chinese ('遇到问题') while the rest of the skill is in English, and the document does not state that the skill is region-specific or that multilingual output is optional. This creates a language/locale policy issue because it imposes a language-specific instruction without user opt-in or documented need.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill tells the agent both to obtain authorization for sensitive operations and to 'never give up' with automatic retries on failure. That conflict can pressure an agent to keep attempting sensitive actions after denial, timeout, or ambiguous approval state, increasing the chance of unauthorized execution or repeated unsafe requests.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The prohibition says 'Abandoning tasks on timeout' is forbidden, implying retries or recovery are required. Later workflow text says to 'wait for confirmation' before execution, so a timeout in getting confirmation could conflict with the never-abandon rule and push execution without approval; this creates contradictory operational intent in the documentation itself.

Content

No source excerpt is available for this finding.