Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Oktk
v2.4.0LLM Token Optimizer - Reduce AI API costs by 60-90%. Compresses CLI outputs (git, docker, kubectl) before sending to GPT-4/Claude. AI auto-learning included. By Buba Draugelis 🇱🇹
⭐ 0· 1k·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The code implements filters/caching/analytics that align with the 'token optimizer' description. However SKILL.md declares node as a required binary while the registry metadata lists no required binaries — an inconsistency. The presence of analytics, cache, and AI-learning subsystems is plausible for this tool but expands scope beyond a minimal 'output compressor' (it persists learning, stats, and cache).
Instruction Scope
SKILL.md instructs users to pipe outputs and source shell aliases; that is expected. But runtime behavior (from implementation files) includes writing logs, stats, cache, and learned-pattern files under the user's home directory (~/.oktk) and reading/writing config there. Those file operations are not emphasized in the top-level SKILL.md metadata and the skill uses environment toggles (e.g., OKTK_DISABLE, OKTK_AI_LEARN) that are not declared in the registry metadata. The analytics system will record sanitized command lines locally by default, which is a privacy risk if you assume no persistent telemetry.
Install Mechanism
There is no install specification in the registry (instruction-only), yet the package contains an installable Node CLI (package.json, bin mapping) and many scripts. No remote downloads or external installers are used. The lack of an explicit install step in the registry vs. the presence of full code is an inconsistency to be aware of, but the code itself does not fetch arbitrary remote artifacts.
Credentials
Registry metadata says no required env vars, but the code reads and respects multiple environment variables (OKTK_DISABLE, OKTK_CACHE_TTL, OKTK_CACHE_DIR, OKTK_LOG_FILE, OKTK_STATS_FILE, OKTK_DEBUG, OKTK_AI_LEARN, OKTK_AI_MODEL, etc.). Analytics is enabled by default unless explicitly disabled. The skill logs sanitized commands and metrics to files in the user's home — this is more privileged than a simple stateless transformer and should be justified/consented to by the user.
Persistence & Privilege
The skill persists cache, analytics logs, stats, and learned patterns to ~/.oktk (and suggests a config file ~/.oktk/config.json). It does not request always:true or system-wide config changes and does not appear to modify other skills. Persistence to the home directory is expected for a tool with caching/analytics/learning, but users should know this creates local files that survive restarts.
What to consider before installing
This skill appears to implement what it claims (compress CLI outputs) but it also stores persistent data and logs locally by default. Things to consider before installing:
- It expects Node and contains a CLI; confirm whether the registry metadata (which omitted node) matches your install path. Ensure you run it with Node >=18 if you use the included scripts.
- By default analytics and caching are enabled and write to ~/.oktk (savings.log, stats.json, cache/, learned/). If you don't want local telemetry, set OKTK_DISABLE=true and/or OKTK_DEBUG=0 before running, or inspect/clean ~/.oktk after testing.
- The analytics sanitizes some secret patterns but may not catch all sensitive data; avoid piping extremely sensitive outputs (full tokens, secrets) through the filter until you review the code and test in an isolated environment.
- The AI-learning feature creates learned-pattern files in your home (~/.oktk/learned) when enabled (OKTK_AI_LEARN=1). If you prefer no learning, keep it disabled.
- Because the registry metadata and SKILL.md differ about required binaries/env, review package.json and scripts locally before trusting an automated install. Test the tool in a disposable environment (or a VM/container) to verify behavior and file writes.
Overall: reasonable functionality but with privacy/persistence surprises and metadata inconsistencies — review and test before enabling in production.Like a lobster shell, security has layers — review code before you run it.
latestvk97abk531ta5t9c8s2f4enek518101b8
License
MIT-0
Free to use, modify, and redistribute. No attribution required.