ClawHub
Back to skill

Security audit

Kimiim

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Kimi group-chat workflow helper, with proportionate chat, file, and workspace-memory behavior for that purpose.

Before installing, understand that this skill may read Kimi group context, list members and messages, send replies or files through kimiim-cli, download shared files, and create workspace-scoped memory/files under .openclaw. Treat downloaded group content and stored memory as untrusted unless you verify it.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill description uses mandatory, expansive trigger language like 'MUST use this skill whenever' for any message related to Kimi group chat, sessions, files, attachments, or multi-agent collaboration. That broad scope can cause the skill to activate in cases where a narrower tool or no tool use would be safer, increasing the chance of unintended reads, downloads, or writes in response to loosely related prompts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill instructs the agent to read and write local workspace memory files and to download remote files, but it does not prominently warn users that using the skill may create or modify local files. In an adversarial group-chat context, that omission can lead to silent persistence of untrusted content in .openclaw/workspace/... or unreviewed file downloads, increasing the risk of data contamination, unsafe follow-on use, or accidental overwrite of local artifacts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.