ClawHub
Back to skill

Security audit

Orchata

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Orchata knowledge-base guide, but users should be careful with its documented delete tools.

Install only for agents that should access and manage Orchata knowledge bases. Treat delete actions as sensitive: confirm the exact space or document ID with the user first, prefer archival where available, and ensure the connected MCP credentials are scoped to the knowledge bases the agent should use.

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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill exposes space deletion through manage_space without warning about the operational consequences, such as loss of access, disruption to knowledge retrieval, or possible broad data impact depending on backend semantics. Because this is an admin-like action embedded in a general-purpose knowledge skill, an agent may invoke it too readily if prompted, especially under adversarial or unclear instructions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill documents a permanent document deletion capability but provides no guidance to require user confirmation, authorization checks, or clear warning about irreversibility. In an agent setting, this increases the chance that a model could delete data based on ambiguous, mistaken, or prompt-injected instructions, causing integrity and availability loss.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.