cli output compression for token savings with rtk

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:25 PM UTC
• Version:
  1.0.0

Overview

This is a disclosed guide for using a CLI output-compression wrapper, with manageable cautions around secrets, retained logs, and verifying the external rtk binary.

Install this only if you trust the rtk binary available in your environment. Treat rtk as an output display layer, not permission to run mutating commands such as commits, pushes, deployments, docker, or kubectl actions. Avoid using environment-variable or log commands around secrets unless you are comfortable with values entering terminal output, model context, local tee logs, or command history.

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 (2)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly advertises `rtk env -f AWS` to display filtered environment variables, which can expose secrets such as access keys, session tokens, or internal endpoints into the model context or terminal logs. In this skill's context, the token-optimization wrapper encourages routine use on verbose commands, making accidental disclosure more likely because users may treat the wrapper as safe-by-default and overlook that environment variables often contain sensitive credentials.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The documentation explicitly introduces `rtk env -f AWS` to surface environment variables but provides no warning that environment variables frequently contain credentials, tokens, and other secrets. In an LLM-agent setting, encouraging environment inspection without strong cautions can cause inadvertent secret disclosure into model context, logs, or chat transcripts.

VirusTotal

65/65 vendors flagged this skill as clean.