Back to skill
Skillv1.0.0

ClawScan security

Unified Find Skills · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignFeb 18, 2026, 4:01 AM
Verdict
Benign
Confidence
high
Model
gpt-5-mini
Summary
The skill is internally consistent with its stated purpose (finding/installing agent skills) and does not request unrelated credentials or persistent privileges.
Guidance
This skill appears to do what it says: search registries and suggest install commands. Before running any suggested install commands (npm install -g, npx skills add, tessl/clawhub installs, or curl pipelines), review the package source and repository links yourself. Global installs and automated flags (-g, -y) install code from the network without prompting—consider installing in a sandbox or with a non-root user, and avoid blindly piping curl to shell. If you want additional assurance, ask the skill to first show the exact repository/URL and README for each candidate skill before performing any install.

Review Dimensions

Purpose & Capability
okThe name/description match the instructions: the SKILL.md only searches three registries (skills.sh, clawhub, tessl.io) and provides install commands. No unrelated secrets, binaries, or system-level accesses are requested.
Instruction Scope
noteInstructions remain within the scope of searching and installing skills, but they recommend running network commands (npx, curl, grep) and installing third‑party CLIs (npm install -g clawhub, tessl CLI). These are expected for this purpose but involve executing code fetched from the network—users should review packages/URLs before running installs.
Install Mechanism
okThis is an instruction-only skill (no install spec). It does not install code itself. It instructs the agent/user to use npx, npm, clawhub, tessl, and curl which are standard mechanisms for discovering/installing skills and are proportional to the task.
Credentials
okNo environment variables, credentials, or config paths are requested. The recommended commands do not attempt to read unrelated system credentials or files.
Persistence & Privilege
okThe skill is not always-enabled and does not request elevated or persistent privileges. It does not modify other skills or system-wide configuration by itself.