Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
querydb-skill
v1.0.0数据库查询 Skill。当用户需要连接 MySQL/PostgreSQL 等数据库执行 SQL 查询、获取测试数据、验证接口返回数据时使用此 Skill。
⭐ 0· 54·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
Name/description, runtime instructions, and the included Python code all implement a DB query/test-case generator for MySQL/Postgres — this is coherent with the stated purpose. However, the skill ships with concrete connection details (IP, database name, username, password, buyer tax number) in the SKILL.md examples and examples use them directly, which is not necessary for delivering the skill and is questionable as a default.
Instruction Scope
SKILL.md instructs the agent to connect to a specific host (10.115.96.247) with a clear-text username/password and to run arbitrary SQL queries against bw_jms_main1. While running queries is expected, the instructions embed and promote using those specific credentials and a taxpayer number (BUYER_TAXNO), increasing risk of unauthorized access or data exposure. The instructions do not instruct exfiltration to external endpoints, but they allow arbitrary SQL execution on that host.
Install Mechanism
No install spec or remote downloads are present; this is an instruction-only skill with a single Python module. That reduces supply-chain risk.
Credentials
The skill declares no required environment variables or credentials in metadata, yet SKILL.md and examples include hardcoded credentials (host, user, password, database) and a specific buyer tax number. Embedding concrete secrets in the skill is disproportionate and sensitive. There are no unrelated third-party credentials requested, but the included secrets themselves are problematic.
Persistence & Privilege
always is false, skill is user-invocable and allows autonomous invocation (default). The skill does not request system-wide config changes or other skills' credentials and does not require persistent installation steps.
Scan Findings in Context
[hardcoded-credentials] unexpected: SKILL.md and examples contain a concrete DB host (10.115.96.247), database (jxindependent0), username (jxindependent), and password (Xj2zCkLJXTkEJ5j) and a taxpayer number (91440606MA4WHN8C8X). These are usable secrets for the skill's purpose but should not be shipped in plaintext in a reusable skill.
What to consider before installing
This skill appears to do what it says (query DBs and generate test cases), but it ships with hardcoded, usable credentials and a taxpayer ID inside SKILL.md and examples. Before installing or running it: (1) treat the embedded host/username/password as sensitive — do not assume they are safe or public; (2) verify with your organization whether that DB and credentials are legitimate and allowed to be used; (3) do not run the skill on networks where access to that host would leak data; (4) remove or replace any hardcoded credentials with environment variables or a secrets manager and update metadata to declare required creds; (5) review and sandbox any SQL you or the skill runs (it permits arbitrary queries); (6) if these credentials are real and were exposed unintentionally, rotate them and audit access. If you cannot validate the provenance of the embedded DB details, avoid using the skill or run it in an isolated test environment only.Like a lobster shell, security has layers — review code before you run it.
latestvk9722wgn7jbpay7g0am5tfqst984dnp9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.