Skillv2.0.1

ClawScan security

Skill Template · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 18, 2026, 10:50 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code and runtime instructions match its stated purpose as a skill/template generator and simple local data CLI; nothing requests unrelated credentials or external network access.
Guidance: This skill appears to do what it says: a local template generator and a tiny data-logging CLI. Before installing or invoking: 1) Note it will create and write to a data directory (~/.local/share/skill-template by default) and append commands/arguments to history.log — don't store secrets or passwords via the tool. 2) It requires bash and python3; ensure those match your environment. 3) The validator/enhancer reads SKILL.md and tips.md in any target directory you give it — avoid validating directories that contain sensitive files unless you intend the skill to read them. 4) Review the included scripts if you want to be extra cautious; they contain no network calls or credential handling.

Review Dimensions

Purpose & Capability: okThe name and description (skill template / scaffold generator and a simple CLI data tool) align with the included files: a small data-logging CLI (scripts/script.sh) and a template/validator generator (scripts/skill-tmpl.sh). Required binaries and env vars declared (none) are consistent with the scripts which only rely on bash and python3.
Instruction Scope: noteInstructions and scripts operate on a local data directory (~/.local/share/skill-template by default) and on a provided target directory when validating/enhancing/publishing templates. They read SKILL.md, tips.md and list scripts/ content in the target. The CLI also writes a history.log that records commands and arguments — this can capture any sensitive strings the user supplies, so avoid adding secrets as entries. No external endpoints or network exfiltration are present in the scripts.
Install Mechanism: okNo install spec (instruction-only with bundled scripts). All code is included in the package; there are no downloads or remote installs. This is low-risk from an install mechanism perspective.
Credentials: noteNo credentials or special environment variables are required. The scripts read SKILL_TEMPLATE_DIR, XDG_DATA_HOME and HOME (standard local configuration) to determine storage location — reasonable for a local CLI. No secrets or unrelated service tokens are requested.
Persistence & Privilege: okalways is false. The skill writes files only under the user's data directory (or alternate SKILL_TEMPLATE_DIR if set) and does not modify other skills or system-wide configs. Autonomous invocation is allowed by default (platform normal) but not combined with elevated privileges.