ClawHub
Back to skill

Security audit

skill-kannaka-constellation

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only monitoring skill for Kannaka constellation/swarm status, with network participation behavior disclosed and no bundled executable code.

Install only if you use the Kannaka constellation tooling. Keep the default local NATS URL or use a NATS server you control, and be aware that joining or listening to a swarm can expose agent identifiers and runtime phase/state information to that bus.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill metadata includes broad auto-activation phrases such as "what's connected" and "all services," which can overlap with normal conversation and cause the skill to trigger unexpectedly. Because the skill exposes commands that inspect infrastructure state and can lead users toward networked swarm actions, accidental activation increases the chance of unnecessary system probing or disclosure of operational details.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documented kannaka swarm join command causes an agent to announce itself and publish phase state onto a NATS bus, which transmits agent identifiers and runtime state over network infrastructure. Although a warning appears later in the document, it is not prominent at the command itself, so users may execute the join action without understanding the privacy, trust-boundary, or external connectivity implications.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.