ClawHub
Back to skill

Security audit

Shell Editing Workflow Helper

Security checks for vulnerabilities and agentic risk

Overview

This skill is a lightweight workflow prompt with no executable code, persistence, credential handling, or data access, though its activation wording is overly broad.

Installers should treat this as a benign prompt helper, but should narrow or disable implicit invocation if they do not want it appearing during ordinary edit, write, shell, or feature requests.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (5)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill description activates on very broad categories like "creative-and-content," "feature," "edit," "write," and "shell," which overlap with many ordinary user requests. This can cause the skill to be invoked outside its intended safety-validation scope, leading to prompt routing confusion, unintended interference with other skills, or inappropriate handling of shell/edit/write tasks that may carry security implications.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The trigger keywords include highly generic terms such as "feature," "edit," "write," "shell," "checked," "existing," "issues," and "found," which are too ambiguous to safely distinguish this skill from normal conversation. In a skill that influences operational workflows around shell and file-modifying actions, accidental activation increases the risk of misrouting sensitive requests and applying the wrong guidance in security-relevant contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The default prompt and description use very broad activation language ('creative-and-content, feature, edit, write, shell, or needs a practical workflow...') without narrow scoping or eligibility checks. In a skill intended to influence edit/write/shell workflows, this can cause unintended routing into a capability that affects sensitive tool-use decisions, increasing the chance of prompt injection propagation, overreach, or unsafe automation in unrelated contexts.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Enabling implicit invocation without clear activation boundaries allows the platform to auto-select this skill in ambiguous situations. Because the skill concerns pre-execution safety checks for edit/write/shell actions, over-invocation or invocation in attacker-shaped contexts could alter tool workflows or inject untrusted guidance at moments involving potentially destructive operations.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The short description and default prompt combine English with Chinese text, which imposes a multilingual presentation choice without stating whether the user can opt into a preferred language. Under the policy, language or locale behavior should not be forced unless choice or justification is provided.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.