ClawHub
Back to skill

Security audit

schoolpass

Security checks for vulnerabilities and agentic risk

Overview

The skill clearly describes access to a parent’s SchoolPass account and requires approval for changes to a child’s dismissal arrangements.

Use a reviewed, pinned package version and protect the configuration containing your password. Keep confirmation set to ask-user, or use refuse if you only need reads; review the child, date, and proposed arrangement before approving changes. Investigation of the scanner’s missing references found an installation target, a remote hostname, a metadata key, and an unbundled test—not additional bundled executable files. This verdict covers the supplied instructions; it does not certify the external package or its claimed safeguards.

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 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

MCP Config Access

High
Category
Agent Snooping
Confidence
95% confidence
Finding

Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Content

Scanner excerpt · SKILL.md (reported line 17)May include surrounding context.

Setup

Add to .mcp.json in your project or ~/.claude/mcp.json:

json
{

The read-oriented manifest exposes tools that modify a child's actual arrival/dismissal arrangements.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The manifest description at L003 scopes the skill to questions and requests to read a SchoolPass parent account. However, L065-L066 expose submission and cancellation of dismissal changes, and L159-L160 explicitly state that these tools change a child's real dismissal. Confirmation gating reduces accidental changes but does not resolve the mismatch between the manifest's stated purpose and the documented write operations.

Content

No source excerpt is available for this finding.