ClawHub
Back to skill

Security audit

Newsletter Creation Curation

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only newsletter planning skill with disclosed approval guidance and no evidence of hidden code, exfiltration, destructive behavior, or privilege-seeking.

Before installing, treat this as a public-content strategy aid: verify claims, avoid sharing confidential company metrics unless you have authority, and add legal or manager review for fintech, HR, employee-led, or regulated content. Users outside India or the US should provide their own market context explicitly.

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 (5)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
```
ADVANTAGES:
✅ No approval needed (publish freely)
✅ Personal voice = authentic
✅ Company brand = personal brand
✅ Can share metrics ("We're at $5M ARR")
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
CHALLENGES:
⚠️ Company may want control over messaging
⚠️ Can't share company metrics without approval
⚠️ Must add "Views are my own" disclaimer
⚠️ Manager needs to be aware
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.

Natural-Language Policy Violations

Low
Confidence
83% confidence
Finding
The playbook instructs users to use Indian companies, rupee pricing, and Indian regulations for the India strategy as fixed defaults. This can amount to forcing a specific locale/language framing unless the user has explicitly opted into that regional context.

Natural-Language Policy Violations

Low
Confidence
82% confidence
Finding
The US strategy section directs the skill to use US companies, dollar pricing, and US regulations as defaults. Without an explicit opt-in or prior confirmation of the user's target market, this creates a locale policy concern by assuming a regional framing.

Natural-Language Policy Violations

Low
Confidence
86% confidence
Finding
The natural-language instructions constrain geography handling to two preset locales, `India-first` and `US-first`, and later workflow steps repeat these fixed settings. This can amount to a locale policy restriction because users outside those locales are not given an explicit choice or fallback behavior.

Static analysis

No suspicious patterns detected.