ClawHub
Back to skill

Security audit

Backup Validator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local OpenClaw backup validator whose sensitive behavior is disclosed and aligned with checking backups before push.

Install only if you want an Indonesian-language helper for validating local OpenClaw backup folders before commit or push. Review any suggested git cleanup command before allowing it, and do not treat PASS as proof of full backup integrity unless you add or separately run hash, permission, and restore checks.

SkillSpector

Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (4)

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The main execution steps describe checks for structure, skill files, secrets, git state, remote sync, compatibility, and metadata, but do not include hash verification, permission validation, or restore testing. However, the 'Common Mistakes', 'Red Flags', and 'Quick Reference' sections state that integrity checks, permission checks, and restore verification are expected, creating a direct contradiction between the documented validator behavior and what the skill actually instructs to do.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The README says to use the skill 'when the conditions in the description are met' but does not state those conditions in this file. For a markdown file, this creates an ambiguous trigger because readers cannot tell from the README when the skill should or should not be invoked.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The skill content is written as imperative operational guidance in Indonesian (for example, 'Gunakan skill ini ketika' and 'Memvalidasi seluruh artefak backup'), and no alternative language or user opt-in is provided. This creates a locale/language constraint in the skill's natural-language interface, which matches the policy category for forced language without user choice.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
- `git status --porcelain` harus kosong setelah commit
- Jika ada modified tetapi belum di-commit:
  - laporkan WARNING
  - minta commit atau `git reset --hard` jika aman
- Jika ada untracked file penting:
  - laporkan WARNING
  - minta `git add` atau sesuaikan `.gitignore`
Confidence
65% confidence
Finding
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:186