Kanban Workflow Export

• Outcome: Pass
• Latest audit: May 28, 2026, 10:13 AM
• Version:
  0.1.4

Overview

This appears to be a legitimate Kanban/project-management automation skill, with meaningful but disclosed ability to read and update work items through your authenticated CLI accounts.

Install only in workspaces where you are comfortable letting the skill act with your existing PM CLI/API permissions. Use least-privilege tokens or scoped CLI sessions, review config/kanban-workflow.json before committing it, and enable cron or automatic progress updates only when periodic external comments and stage changes are acceptable.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (5)

Missing User Warnings

Low

• Confidence: 84% confidence
• Finding: The setup flow states that it writes config/kanban-workflow.json, but the warning is easy to miss and does not clearly frame this as a filesystem side effect before execution. Undisclosed local writes reduce informed consent and can surprise users, especially in automation contexts where working-directory file changes may affect other tools or leak configuration into repos.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The skill advertises automatic progress-update comments every 5 minutes for in-progress tasks but does not provide a prominent warning that this causes outbound writes to external PM systems. In agentic environments, automatic posting can disclose internal status, generate noisy or unintended communications, and act through the user's authenticated CLI context without sufficiently explicit consent.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The requirements explicitly mandate automatic progress-update comments every 5 minutes while a task is in progress, which creates recurring external writes without a per-action user confirmation or even a warning that autonomous posting will continue in the background. In an agent skill, this can leak sensitive work details, spam external systems, and cause unintended side effects long after the initiating command.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The requirements for `show` and `next` explicitly include private/internal comments where supported, but provide no privacy guardrails, minimization, or warning before surfacing that data to the agent or user. This can expose sensitive internal discussions, credentials, incident details, or HR/legal content beyond the minimum needed for task execution.

Missing User Warnings

Medium

• Confidence: 79% confidence
• Finding: This function automatically posts comments to external work items without any built-in confirmation, disclosure, or explicit opt-in mechanism. In an agentic workflow skill that integrates with PM systems, that creates a real risk of unintended external side effects, noisy spam, or disclosure of incorrect/generated status text to collaborators, especially when `getMessage` is agent-supplied.

VirusTotal

66/66 vendors flagged this skill as clean.