ClawHub
Back to skill

Security audit

press-release-writer

Security checks for vulnerabilities and agentic risk

Overview

The skill needs review because its advertised compliance checker misses obvious risky claims and reports “No flags.”

Use this for drafting only and retain legal and finance review. Do not treat “No flags” as clearance; the matching logic needs correction and testing before relying on it. The optional Windows application is unnecessary for the bundled workflow and is outside this assessment.

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
98% confidence
Finding

The reported use of literal substring matching for regular-expression patterns would cause risk checks to miss problematic claims, creating false assurance in a workflow explicitly advertised as checking advertising and securities language. Named legal and finance reviewers reduce the risk, but users could still rely on incomplete flags when preparing consequential announcements. The script itself was not supplied, so this assessment depends on the static analyzer's implementation report; lack of draft generation alone is not a security vulnerability because the agent could perform that part.

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.

The advertised compliance check treats regex rules as literal strings, missing the risk language it claims to flag.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The checker uses substring membership (pattern in low) instead of regular-expression matching, so ordinary risky phrases such as "risk-free", "best product", and "we expect" do not match the configured regex strings and can produce a misleading "No flags" report. This is particularly consequential for a skill advertised as checking advertising and securities risks, because users may rely on the report when approving public statements; the human-review disclaimer limits but does not eliminate that risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.