ClawHub
Back to skill

Security audit

Agent-Friendly CLI

Security checks for vulnerabilities and agentic risk

Overview

This is a guidance-only CLI design skill with no executable code, hidden behavior, or persistence.

Installers should treat this as CLI design guidance. When applying its advice to real production or destructive commands, keep explicit safeguards such as dry-run, clear scoping, audit logs, and safe defaults even if an agent-facing --yes or --force option exists.

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
  • 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 (3)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% 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 50)May include surrounding context.

md
Options:
  --env     Target environment (staging, production)
  --tag     Image tag (default: latest)
  --force   Skip confirmation

Examples:
  mycli deploy --env staging

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% 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 111)May include surrounding context.

No changes made.

text

### 8. --yes / --force to Skip Confirmations

Humans get "are you sure?" — agents pass `--yes`.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% 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 172)May include surrounding context.

md
| 5 | Fast failures | Do missing args error immediately with usage hint? |
| 6 | Idempotency | Is running the same command twice safe? |
| 7 | Dry run | Do destructive commands support `--dry-run`? |
| 8 | Skip prompts | Is there `--yes` / `--force` for confirmations? |
| 9 | Predictable structure | Is command naming consistent (resource + verb)? |
| 10 | Structured output | Does success output include actionable data? |
| 11 | JSON output | Is `--json` available for machine parsing? |