ClawHub
Back to skill

Security audit

Daily Wisdom

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed daily storytelling prompt that uses web research, a local history file, and user-configured scheduled delivery without hidden code or unexpected privileges.

Before installing, confirm the cron schedule and destination channel, because enabled runs will generate and send daily content automatically. Use a dedicated history file and disable or remove the cron job when you no longer want recurring delivery.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The README promotes automated daily delivery to WhatsApp, Telegram, Slack, and Discord and later instructs the user to set up a cron job, but it does not explicitly warn that this causes unattended outbound messages to third-party services. For a markdown skill description, this is a behavior affecting privacy and external data transmission that should be disclosed clearly to users.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The manifest and overview describe the skill as generating a new story from a listed source pool and tracking history to avoid repeats. However, the core prompt instructions explicitly require using web search to verify quotes, facts, and details before writing, which expands behavior beyond simple local content generation and cron delivery.

Natural-Language Policy Violations

Low
Confidence
77% confidence
Finding
The file presents a quoted passage in Japanese, which can be interpreted as imposing a specific language/locale in the skill content without any opt-in or alternative for users who may not read Japanese. Although an English translation follows, the primary quoted content is still language-specific and not explicitly justified as a locale-bound skill requirement.

Natural-Language Policy Violations

Low
Confidence
87% confidence
Finding
The file includes a quotation in Devanagari script with no indication that the skill supports user language preference or provides an opt-in for non-English content. Because language/locale policy violations apply to all file types, presenting untranslated source-language text as part of the main content can be a locale-choice issue when no user preference is offered.

Static analysis

No suspicious patterns detected.