ClawHub
Back to skill

Security audit

ctexcel-usage-skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated benign with high confidence: the scanners report no AIG findings and only low-level concerns about user confirmation and advertised allowance checks.

The verdict is benign, and the scan reports complete coverage of the four listed components. Before installing, consider whether the reported gap in voice and SMS allowance reporting matters for your use; benign does not mean risk-free.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 26)May include surrounding context.

md
## Safety And Consent

- Never ask the user to paste a password. Login is SMS-only.
- Ask for the phone number and SMS code only when a login is actually needed.
- Do not echo the full phone number, the verification code, the session cookie,
  or the contents of the state file. Use the CLI's masked output (`07****84`).

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/session.md (reported line 32)May include surrounding context.

}

text

No password and no verification code are ever written.

`pending_login` bridges the two login commands. It holds the anonymous session
created by `--request-sms` and expires after 10 minutes. `last_sms_at` enforces

The manifest promises voice and SMS allowance checks, but the documented CLI output excludes both.

Low
Category
Excessive Agency
Confidence
97% confidence
Finding

The manifest describes checking remaining data, voice, and SMS allowances. Lines L107–L110 explicitly state that SMS and voice fields returned by the carrier are not reported, so the documented CLI output cannot fulfill the advertised voice and SMS checks.

Content

No source excerpt is available for this finding.