ClawHub
Back to skill

Security audit

Joan Workflow

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill that guides users through Joan workspace commands, with the main caution being to review what gets written into local context files before sharing or using it.

Install only if you intend to use Joan for workspace knowledge and task management. Before running context generation or push commands, review which pods are tracked and what will be written to CLAUDE.md or shared with the Joan server.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructs users to run joan context claude to inject workspace context into CLAUDE.md, but it does not warn that this modifies a local file and may copy workspace-derived content into the assistant context. In a knowledge-management workflow, that can unintentionally expose sensitive internal documentation, broaden prompt-injection exposure from workspace content, or cause users to trust generated context without review.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.