Tene CLI — Local-First Secrets

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.3

Overview

This is a documentation-only skill for using the Tene secret manager, and its sensitive behavior is disclosed and aligned with that purpose.

Install this only if you intend to use Tene for project secrets. Review the exact commands before allowing an agent to run them, especially imports, deletes, deploys, and production runs. Prefer verifying imported key names and keeping a temporary backup before deleting any .env files, and do not paste secret values, master passwords, or recovery phrases into chat.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The activation criteria are extremely broad for a secret-management skill, causing it to trigger on generic mentions of secrets, tokens, environment variables, or commands that may not actually involve tene. This can redirect unrelated workflows into this skill, increasing the chance the agent requests, handles, or transforms sensitive material unnecessarily.

Vague Triggers

Medium

• Confidence: 92% confidence
• Finding: The example trigger phrases include ordinary requests like storing a key or injecting env vars, which are common across many tools and contexts. Ambiguous invocation boundaries make accidental activation likely, exposing users to tool-specific guidance and secret-handling flows when a different secret store or a non-secret task may be intended.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The example directs deletion of the plaintext `.env` file immediately after import without an explicit confirmation or backup-validation step. In an agentic context, that can cause irreversible data loss if import failed, imported incomplete values, or targeted the wrong file, especially because the agent is shown executing `rm .env` non-interactively.

VirusTotal

67/67 vendors flagged this skill as clean.