ClawHub
Back to skill

Security audit

Linear Flow Cli Free

Security checks across malware telemetry and agentic risk

Overview

This Linear skill is mostly coherent, but it needs review because it can change live project data and includes risky token-display guidance with broad activation language.

Install only if you want an agent to operate Linear through the CLI. Keep Linear API keys out of shared logs and chats, avoid printing tokens unless absolutely necessary, add .linear.toml to .gitignore, and require explicit confirmation before creating or updating Linear issues, labels, comments, or raw API calls.

SkillSpector

By NVIDIA
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
Confidence
81% confidence
Finding
The skill's activation scope is described with a very broad keyword set spanning generic project-management and team-collaboration terms, which increases the chance the agent invokes it in situations the user did not explicitly intend. Because the skill has `exec` access and supports write operations against Linear, overbroad activation can cause unintended reads, task creation, or status changes in a live project-management system.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill explicitly instructs users to view the current token with `linear auth token` and notes persistent credential storage in `.linear.toml`, but it does not prominently warn that these credentials may be exposed through terminal output, logs, screenshots, shell history, or insecure filesystem permissions. In an agent context, encouraging token display materially raises the risk of secret disclosure because command output may be captured or relayed automatically.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.