Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

代码质量分析系统

v1.2.0

完整的代码质量分析系统，包含前后端服务和数据库。支持周/月维度分析、AI代码审查、Teams消息通知、邮件报告。触发场景：(1) 用户要求进行代码质量分析 (2) 生成周报/月报 (3) 统计代码变更 (4) 分析分支提交 (5) 同步分析数据到数据库。

⭐ 1· 126·1 current·1 all-time

by@fightingdao

MIT-0

Download zip

LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.

Security Scan

VirusTotal

Pending

View report →

OpenClaw

Suspicious

medium confidence

Purpose & Capability

Name/description align with the included code and scripts (analysis, AI review, DB sync, Teams/email notifications). However the registry metadata claims no required env vars/credentials while README and multiple scripts clearly require database credentials (DATABASE_URL / Prisma), SMTP credentials, and Teams webhook/secret. The skill therefore underdeclares sensitive requirements that are legitimately needed for its functionality but are not declared in metadata.

Instruction Scope

SKILL.md and README instruct the agent to clone repos, run npm install, initialize a database, read config files from ~/.openclaw and workspace paths, run many git commands (fetch, log, diff), and send notifications to external Teams webhooks and SMTP recipients. Those actions are within the stated purpose, but the runtime instructions also enforce a policy to 'generate at least 1 code issue for every project' (i.e. fabricate issues when none are detected), which is an integrity/accuracy concern and unexpected for an analysis tool.

ℹ

Install Mechanism

Skill is instruction-only (no install spec) which is low platform-install risk. The README asks to git-clone backend/frontend from GitHub and run npm install and Prisma migrations — network downloads and package installs are required at runtime. The external sources referenced (GitHub repos) are normal but the skill will pull and run third-party code and Node packages on your machine.

Credentials

Although registry metadata lists no required env vars, the code and docs require sensitive configuration: DATABASE_URL / backend .env (Postgres + JWT_SECRET), SMTP credentials (host/user/pass or auth code), Teams webhook URL and secret, and config.json with codebaseDir and project repository locations. That is proportionate to a full system, but the omission from metadata is misleading and increases the chance a user will supply secrets without realizing scope. The scripts also read files under HOME (~/.openclaw/...), so they need filesystem access to your workspaces and repos.

ℹ

Persistence & Privilege

always:false (good). The skill can be invoked autonomously (platform default). Combined with DB write operations (Prisma) and notification sending (Teams/SMTP), an autonomously-invoked skill could write data to your DB, email recipients, or post to external webhooks; this raises operational risk but is expected for this type of tool. No evidence the skill attempts to modify other skills or system-wide agent config.

What to consider before installing

This skill implements a full analysis pipeline but omits declaring required credentials in its metadata and enforces a rule to generate at least one issue per project even when none exist. Before installing or running: 1) Review the code locally (especially scripts/*.js) — they run many git commands, read config files under ~/.openclaw, and use Prisma to write to a database. 2) Don’t provide real DB/SMTP/Teams credentials until you’ve audited the backend and confirmed where data will be sent. 3) Consider running it in an isolated environment (throwaway VM/container) with sample repos and a test database to observe behavior. 4) If you need truthful analysis results, ask the author to remove or document the enforced 'generate at least one issue' behavior. 5) If you proceed, limit network access or use test webhooks/email accounts to avoid accidental data leak or noisy notifications.

✗

scripts/analyze-code-v2.js:144