Oktk

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill is coherent for compressing command output, but it wraps shell commands and stores local cache/analytics, so users should enable it deliberately.

This appears safe for its stated purpose if you treat it as a command wrapper. Install/use it only if you are comfortable with it executing commands you pass to it and storing local cache/statistics under ~/.oktk. Avoid using it on secrets or highly sensitive outputs, use --raw when details matter, and only enable shell aliases if you want persistent auto-filtering.

Findings (5)

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.

What this means

Using oktk as a wrapper can run real terminal commands, not just compress already-produced output.

Why it was flagged

The CLI can execute a command string through the shell before filtering its output. This matches the wrapper purpose, but shell execution is high-impact if the agent or user passes unsafe commands.

Skill content
const stdout = execSync(command, {
Recommendation

Use it only with commands you intentionally want to run; prefer explicit pipe mode for sensitive commands and review destructive commands before execution.

What this means

Some command output may be retained locally after the command finishes.

Why it was flagged

Filtered results are cached locally under the user's home directory. This supports the stated performance/cost-saving purpose, but command output may remain on disk for the cache TTL.

Skill content
this.cacheDir = options.cacheDir || process.env.OKTK_CACHE_DIR || path.join(process.env.HOME, '.oktk', 'cache');
Recommendation

Avoid filtering highly sensitive output, set OKTK_CACHE_TTL appropriately, or clear/disable the cache when needed.

What this means

Local analytics files may reveal recent command names, file paths, URLs, or project activity even with basic secret redaction.

Why it was flagged

The analytics module records local savings logs and sanitized command names. This is purpose-aligned for reporting savings, but it creates a local command history.

Skill content
this.logFile = options.logFile || process.env.OKTK_LOG_FILE || path.join(process.env.HOME, '.oktk', 'savings.log');
Recommendation

Disable tracking with configuration if command history is sensitive, and periodically remove ~/.oktk logs if desired.

What this means

Future sessions may use compressed command output through aliases unless the shell startup line is removed.

Why it was flagged

The skill suggests adding aliases to shell startup files for automatic filtering. This is user-directed and disclosed, but it changes future shell behavior.

Skill content
source ~/.openclaw/workspace/skills/oktk/scripts/oktk-aliases.sh
Recommendation

Only source the alias file if you want persistent auto-filtering; keep the --raw bypass available and remove the shellrc line to disable.

What this means

The skill may not work unless Node is present, and automated installers may not validate that requirement.

Why it was flagged

The registry metadata under-declares runtime needs compared with SKILL.md's Node requirement and package.json's CLI bin. This is an install/provenance clarity issue, not evidence of malicious behavior.

Skill content
Required binaries (all must exist): none
Recommendation

Confirm Node.js is installed and review the repository/source before relying on the global CLI.