ClawHub
Back to skill

Security audit

cairn

Security checks across malware telemetry and agentic risk

Overview

Cairn is a disclosed local indexing and retrieval skill that reads content the user chooses to ingest and stores it in a local SQLite index.

Install only if you are comfortable with selected local files being indexed into ~/.cairn and later retrievable by agents you connect to the MCP server. For sensitive repositories, set CAIRN_ALLOWED_ROOTS, use a separate dbPath, and enable CAIRN_OFFLINE when you need to prevent web fetches and model downloads.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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

Medium
Confidence
84% confidence
Finding
The doc-extraction pass sends local markdown content and code-entity context to a chat model, which can disclose sensitive local repository information to a model backend. Even if the deployment is intended to be local, this file does not enforce locality or require explicit opt-in before transmitting content, so misconfiguration could turn indexing into unintended data exposure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.