ClawHub
Back to skill

Security audit

keyword-research-assistant

Security checks for vulnerabilities and agentic risk

Overview

The bundled workflow performs limited local processing, but its ranking calculation needs correction and the optional Windows app requires separate review.

Use a dedicated output folder because ranking.md is overwritten. Correct and validate the scoring before relying on its recommendations: harder keywords currently rank higher when other inputs match. The scanner's scoring concerns are supported as reliability defects; missing permission declarations do not establish overbroad access in this limited script. The optional Windows executable is unnecessary for this workflow and is outside this verdict.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The script largely implements the declared keyword-ranking workflow, but its difficulty calculation materially undermines the stated prioritization purpose. Although the criterion is labeled 'difficulty scored inverse,' the code applies value * weight without inversion and ranks larger totals first. Local CSV reading and report writing are ordinary supporting behavior; no unrelated access or undeclared external capabilities are evident.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_write.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Keyword difficulty increases the ranking score despite being documented as inversely scored.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

Difficulty is multiplied by a positive weight without inversion, and results are sorted descending, so increasing difficulty improves a keyword's rank despite the documented inverse-scoring requirement. This is a scoring-integrity defect that can misdirect quarterly SEO priorities and waste content resources, but the supplied code shows no associated code execution, data disclosure, or privilege escalation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.