Bỏ qua tới nội dung
TakoAPI
Quay lại kỹ năng

skill-safety-checker

bởi runeweaverstudios

Coding Agents & IDEs
341Lượt tải do clawskills.sh công bố
0Lượt xem
0 lượt thíchĐăng nhập để thích kỹ năng này
—Chưa có đánh giá
Đánh giá của bạn

Tóm tắt

Runs VirusTotal-style security checks on OpenClaw/Cursor skills before install, including remote code execution (RCE)

Mô tả

Kỹ năng này làm gì

Runs structured security checks on skills from registries or shared sources before installation. Covers RCE patterns, malicious code, credential misuse, and mismatches between a registry listing and the skill's SKILL.md. Produces a 10-point report with a Benign or Suspicious verdict.

Rather than ad-hoc judgment, it applies a fixed 10-point checklist covering RCE, malicious code, credential scope, and registry consistency, always ending in a single clear verdict.

Khi nào nên dùng

  • Checking a ClawHub skill for safety before installing it
  • Reviewing a skill that requests OAuth or API key access
  • Auditing all downloaded skills in a local skills directory
  • Evaluating a skill shared via URL or file before granting credentials
  • Self-checking an authored skill before publishing to a registry

Quy trình ví dụ

ĐẦU VÀO

User asks: 'Run a security check on the gog-cli skill from ClawHub'

AGENT
  1. 1

    Load the skill's SKILL.md and registry listing

  2. 2

    Check purpose vs capability alignment

  3. 3

    Scan for RCE patterns such as curl|sh, eval of remote input, and unvalidated command construction

  4. 4

    Scan for malicious code signs including obfuscation, exfiltration, secret reads, and backdoors

  5. 5

    Verify credential declarations and install mechanism consistency between registry and SKILL.md

ĐẦU RA

10-point security report with a Benign or Suspicious verdict and 'What to consider before installing' guidance

Cài đặt

clawhub install runeweaverstudios/skill-safety-checker

Chạy lệnh CLI, hoặc dán prompt vào Claude Code, Cursor, Codex hay coding agent bạn đang dùng.

Liên kết