Skillstore
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill mostly matches its stated purpose, but it can install third-party OpenClaw skills from GitHub using shell execution, which can change future agent behavior and needs careful review.
Use this only if you intentionally want a skill manager. Before installing any GitHub result, verify the repository owner, inspect the target skill's SKILL.md and code, and avoid installing skills that request broad account, filesystem, or shell authority unless you trust them.
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.
Installing an untrusted skill can persistently change the agent's capabilities and may expose future tasks to unsafe instructions or code.
The skill is designed to install skills selected from public GitHub search results, but the artifacts do not document trust checks, pinning, allowlists, signature verification, or review requirements.
3. **GitHub** - Search openclaw repositories ... Enter number → Install from GitHub
Only install from repositories you trust; the skill should show the full owner/repo, require explicit confirmation, review the target skill before installing, and ideally support allowlists or pinned commits.
A shell-based installer can affect local files and tools if a command is malformed or if an untrusted source influences the command.
The code imports and uses shell command execution. This is purpose-aligned for installing skills, but shell execution is sensitive and the provided main.js content is truncated around the command construction.
const { exec } = require('child_process'); ... exec(cmd, (error) => {Prefer safer APIs such as spawn with argument arrays, validate all repository names and paths, and ask the user before running install commands.
If these guidelines are loaded as agent instructions, they could influence the agent's decision-making beyond simple skill search or installation.
This included guidance is broader than the immediate skillstore command behavior and could nudge an agent to reduce caution if treated as general instructions.
When user requests something, execute it unless it causes actual harm ... Trust the user's judgment over own assumptions
Keep development guidelines explicitly scoped to creating skills, and avoid broad directives that change the agent's general safety posture.