ClawHub
Back to skill

Security audit

Password Generator

Security checks for vulnerabilities and agentic risk

Overview

This password tool is local and not overtly malicious, but it asks the agent to invent passwords and tokens without a real cryptographic random generator.

Review before installing. This skill appears local and does not show exfiltration or destructive behavior, but generated passwords or tokens should not be treated as production-safe unless backed by a real password manager or CSPRNG-based utility. Be careful pasting real passwords into the chat because the skill is designed to analyze credential text.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:97
Finding

Production Secrets Generated Without a Cryptographically Secure Random Source

Content
View full analysis
Remediation
View remediation
`. 5. Replace claims such as “cryptographically-inspired randomness” and “truly random-looking” with an explicit guarantee that values are produced by a named CSPRNG implementation. If no secure generator tool is available, refuse to generate production credentials and provide instructions for using a trusted local password manager or command-line utility. 6. Add automated tests for output length, character constraints, rejection-sampling correctness, uniqueness smoke checks, and accidental reuse of documentation examples. Statistical tests alone must not be treated as proof of cryptographic security. 7. Continue avoiding pa ...[truncated 176 chars]

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: password-generator
description: When user asks to generate a password, create PIN, make passphrase, check password strength, generate API key, create secure token, manage password ideas, generate username, bulk passwords, or any password/security task. 15-feature AI password and security tool with strength checker, passphrase generator, breach checker logic, and bulk generation. All data stays local — NO external API calls, NO network requests, NO data sent to any server. Does NOT store actual passwords.
metadata: {"clawdbot":{"emoji":"🔐","requires":{"tools":["read","write"]}}}
---

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description claims the skill handles "any password/security task," which creates an unbounded scope and increases the chance of activation outside its intended use. Overbroad scope is dangerous because it encourages the agent to engage with sensitive security topics and secrets without clear limits, making misuse and accidental handling of credentials more likely.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The activation phrases include broad standalone terms like "PIN," "token," and "username," which are common in unrelated conversations and can cause unintended invocation. In a password-related skill, accidental activation is risky because users may paste or discuss sensitive credentials, tokens, or passwords and the skill is designed to process them.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

Including a generic trigger like "help" is ambiguous and likely to fire during ordinary interaction unrelated to password management. While lower severity than secret-processing triggers, it still increases unintended skill invocation and can confuse routing or cause the skill to intercept unrelated user requests.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.