ClawHub
Back to skill

Security audit

account-research-brief

Security checks for vulnerabilities and agentic risk

Overview

The bundled skill has limited local access, but its reports need manual source checks and its supplied template does not match the script.

Treat the output as a draft: verify every claim, source URL, and checked date before sharing. The script does not correctly consume the supplied YAML template. Choose an output directory where replacing digest.md is acceptable. The optional Windows app is unnecessary for this workflow and requires a separate trust decision.

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 code supports the brief's section structure, but its implemented purpose is a local notes-to-Markdown formatter. It copies matching section content without factual verification, source validation, or date checks, and will output unsourced facts unchanged. Printing guardrails about sourcing does not implement the description's guarantee that every fact is tied to a source and checked date. No network access or unrelated malicious capability 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 brief presents input notes as facts without requiring source URLs or checked dates.

Medium
Category
Excessive Agency
Confidence
98% confidence
Finding

The script copies input notes into report sections, including key_facts, without requiring a source URL or checked date; the appended guardrails do not enforce those requirements. Untrusted or inaccurate notes can therefore appear as factual account research, potentially misleading deal decisions. This integrity gap is especially relevant because the skill explicitly promises sourced, date-checked facts, although the script itself does not execute the notes or transmit them externally.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.