local-skill-manager

ReviewAudited by ClawScan on May 1, 2026.

Overview

This skill appears purpose-aligned for managing local skills, but users should notice that it can create and permanently delete local skill directories.

This skill is reasonable for managing local skills. Before installing, be aware that it can permanently delete skill folders; prefer dry-run mode first and do not use --force unless you are sure. Also verify the PyYAML dependency source/version if supply-chain hygiene is important to you.

Findings (3)

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

If used incorrectly, this could remove local skills and change how the user's agent behaves.

Why it was flagged

The skill includes a script that can permanently remove a local skill directory, and it supports a force option that skips the confirmation prompt.

Skill content
force_flag = '--force' in args ... shutil.rmtree(target_path)
Recommendation

Use --dry-run first, avoid --force unless the user explicitly requested deletion, and keep backups of important local skills.

What this means

Installing dependencies can introduce supply-chain risk if the package source or version changes unexpectedly.

Why it was flagged

The skill asks users to install a Python dependency for normal operation; the dependency is purpose-aligned but unpinned and not represented as an install spec.

Skill content
pip install -r requirements.txt ... `requirements.txt` 包含:`pyyaml`
Recommendation

Install dependencies from a trusted package index and consider pinning PyYAML to a known-good version.

What this means

Names, versions, and descriptions of installed local skills may be exposed to the agent during listing, and descriptions from other skills should be treated as data rather than instructions.

Why it was flagged

The listing script reads descriptions from other local SKILL.md files and prints them into the agent context.

Skill content
description = meta.get('description', description) ... print(f"  {skill['description'][:100]}...")
Recommendation

Review listed skill metadata as informational output only, especially if some installed skills come from untrusted sources.