ClawHub
Back to skill

Security audit

Reactivation Specialist

Security checks for vulnerabilities and agentic risk

Overview

This skill is a scoped email reactivation planning guide, not an automatic tool for changing subscriber records.

Before installing, be aware that the skill can work with subscriber engagement and consent-related data. Use it for planning and review its proposed suppression or re-consent rules before applying them in an email platform.

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 (1)

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 68)May include surrounding context.

md
## Decision Gates

- **Stop and ask** — only when the no-engagement window is genuinely unknowable and cannot be inferred (e.g., "win back my dead subscribers" with no recency data available and no stated definition of "dead"), or when there is no consent record on file to check against, which is a **NEEDS_INPUT** for S2 the registry owns — do not design a suppression rule against subjects whose lawful basis is unknown. Present the numbered options (which recency window, which suppression policy) with their outcomes rather than guessing.
- **Continue silently** — do not stop for: a missing ESP engagement export (design the ladder from the stated window, mark cohort-size findings Estimated and proceed); whether an incentive exists (design the ladder with the incentive step marked optional / conditional); optional GA4 return-activity data absent (use last-open/last-click recency alone).

## Save Results

Static analysis

No suspicious patterns detected.