RTK Integration

Security checks across static analysis, malware telemetry, and agentic risk

Overview

The skill has a clear token-saving purpose, but it tells the agent to run an automatic installer that downloads and executes an unpinned remote script and persistently changes shell behavior.

Review the installer before running it. If you install RTK, prefer a pinned release or trusted package manager, confirm any shell profile changes, and use raw shell commands when you need complete output for important operations.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

Installing the skill could run remote code controlled outside the skill package, which may change over time.

Why it was flagged

The installer executes a remote script directly from the master branch without a pinned version, checksum, or review step, so the code run at install time can change independently of the reviewed skill artifacts.

Skill content
curl -fsSL https://raw.githubusercontent.com/rtk-ai/rtk/refs/heads/master/install.sh | sh
Recommendation

Avoid automatic install from an unpinned branch. Prefer a pinned release, package-manager install, or a reviewed script with checksum/signature verification.

#
ASI05: Unexpected Code Execution
Medium
What this means

A user may expect a documentation-only skill, but the instructions encourage immediate code execution during installation.

Why it was flagged

The skill is marked as instruction-only with no install spec, but its instructions ask for immediate execution of a local setup script that may install software and execute a remote installer.

Skill content
When this skill is installed, **immediately run the setup script** to install RTK if not already present: `bash skills/rtk-integration/scripts/rtk-setup.sh`
Recommendation

Require explicit user confirmation before running setup, and declare the install mechanism and required tools in metadata or install specs.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

The agent may see summarized command output rather than full output, including for git commits, pushes, installs, tests, and Docker commands.

Why it was flagged

The wrapper is purpose-aligned, but it covers broad and sometimes mutating shell commands while intentionally compressing output, which can reduce visibility into important command details.

Skill content
When executing shell commands, **always prefer the RTK version** over the raw command... `rtk git add .` ... `rtk git commit -m "msg"` ... `rtk git push`
Recommendation

Use RTK for routine output reduction, but request raw commands/output for sensitive, destructive, or unclear operations.

#
ASI10: Rogue Agents
Low
What this means

RTK may remain active and available in future shell sessions after the skill setup runs.

Why it was flagged

The setup persists RTK on the user's PATH by editing shell profile files, so the change remains after the current session.

Skill content
echo "$PATH_LINE" >> "$PROFILE"
Recommendation

Review profile-file changes after installation and remove them if you do not want persistent RTK availability.