ClawHub
Back to skill

Security audit

market-research-brief

Security checks for vulnerabilities and agentic risk

Overview

The skill uses proportionate local file access, but its helper provides fewer research checks than advertised.

Manually review sources, word limits, and respondent privacy before sharing a brief; the helper does not verify these requirements. Choose an output directory where replacing the report is acceptable. Treat the optional unsigned [Windows app](https://github.com/rapidworker1/Multiple-AI-models-Toolkit) as a separate software decision; this verdict does not assess its executable.

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
99% confidence
Finding

The code's primary output is materially different from the promised brief: it reports whether predefined fields are present and within character limits, without reproducing or synthesizing their content. It also does not implement per-number sourcing. Reading the input file and writing a local Markdown file are ordinary supporting operations, and no unrelated access or hidden behavior is 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.

The script produces a field-length checklist rather than the promised research brief with word limits and a source for every number.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

At L095–L107, field contents are replaced with character-count status lines, so the output omits the actual research narrative and checks characters instead of the manifest's promised word limits. L116–L118 merely append source-related reminders; the code neither associates numbers with sources nor checks their presence, and the input filename recorded at L092 is not per-number sourcing.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.