ClawHub
Back to skill

Security audit

File Management

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward file-management skill that teaches common local filesystem commands and explicitly requires confirmation for write or destructive actions.

Install only if you want the agent to help with local file operations. Before allowing any move, copy, archive, or delete command, check the exact path and command, and prefer Trash or backups for deletion.

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
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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 (6)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

The skill includes a direct example of permanent deletion using 'rm /path/to/file.txt' and labels it as a last resort. Even with surrounding guidance to confirm first, including a ready-to-run destructive command in a general-purpose skill increases the chance of unsafe execution, path substitution mistakes, or agent misuse when handling user-controlled file paths.

Content

Scanner excerpt · SKILL.md (reported line 129)May include surrounding context.

Preferred on macOS (goes to Trash):

trash /path/to/file.txt

Or rm as last resort:

rm /path/to/file.txt

text

### Batch Rename (pure bash)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% 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).

Content

Scanner excerpt · SKILL.md (reported line 173)May include surrounding context.

md
## Anti-Patterns (Never Do)

- ❌ `rm -rf /` or any recursive delete without confirming
- ❌ `chmod -R 777` or permission changes that break security
- ❌ Executing downloaded scripts directly without review
- ❌ Accessing paths outside user's home without asking

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% 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).

Content

Scanner excerpt · SKILL.md (reported line 173)May include surrounding context.

md
## Anti-Patterns (Never Do)

- ❌ `rm -rf /` or any recursive delete without confirming
- ❌ `chmod -R 777` or permission changes that break security
- ❌ Executing downloaded scripts directly without review
- ❌ Accessing paths outside user's home without asking

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% 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).

Content

Scanner excerpt · SKILL.md (reported line 174)May include surrounding context.

md
## Anti-Patterns (Never Do)

- ❌ `rm -rf /` or any recursive delete without confirming
- ❌ `chmod -R 777` or permission changes that break security
- ❌ Executing downloaded scripts directly without review
- ❌ Accessing paths outside user's home without asking
- ❌ `sudo` operations unless explicitly requested

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 · SKILL.md (reported line 160)May include surrounding context.

md
## Permission Model

**Read-only operations** — safe to execute without asking:
`ls`, `find`, `grep`, `cat`, `head`, `tail`, `du`, `stat`, `file`, `mdls`

**Write operations** — always confirm before executing:

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 · SKILL.md (reported line 176)May include surrounding context.

md
## Permission Model

**Read-only operations** — safe to execute without asking:
`ls`, `find`, `grep`, `cat`, `head`, `tail`, `du`, `stat`, `file`, `mdls`

**Write operations** — always confirm before executing:

Static analysis

No suspicious patterns detected.