sql-assistant
Verdict safe, but 1 finding (1 medium), mainly command injection. Most recent scan 2026-02-07.
The sql-assistant skill is a comprehensive SQL query assistant that provides educational content about SQL operations, optimization, and troubleshooting. The primary security concern is that it teaches SQL query construction without adequate coverage of SQL injection prevention, which could lead users to implement unsafe patterns. The skill itself does not execute queries or access databases directly, limiting the immediate risk. However, the absence of security guidance in an SQL-focused skill is a notable gap. Additionally, the skill has incomplete packaging with multiple referenced files missing, and lacks optional metadata fields for transparency. No evidence of data exfiltration, prompt injection, credential theft, or malicious code execution was found. The skill appears to be a legitimate educational tool with room for improvement in security guidance and completeness.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
SkillSpector found no issues but could not fully inspect the package, usually because SKILL.md refers to a file that is not present, so it fails closed to CAUTION instead of SAFE. Read it as "not fully checked", not as a finding. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Comprehensive SQL query assistant for database operations, optimization, and troubleshooting. Use when Codex needs to write, debug, optimize, or explain SQL queries; analyze database schemas; or help with SQL-related tasks including joins, subqueries, aggregations, and performance tuning. Supports MySQL, PostgreSQL, SQLite, and other SQL dialects.
- Package
- 4 files, 33 KB
- SKILL.md
- 1538 words, 12+ sections
- Code
- None. Instructions only.
SKILL.md covers: SQL Assistant · Overview · Core Capabilities · 1. Query Writing · 2. Query Debugging · 3. Query Optimization · 4. Query Explanation · 5. Database Schema Design · Quick Start · SQL Dialect Support · MySQL / MariaDB · PostgreSQL
Findings
MEDIUM SQL Injection Risk - Unvalidated User Input in Query Examples Command Injectionllm
The skill provides extensive SQL query examples and templates without any guidance on parameterized queries or input validation. The instruction body shows direct string concatenation patterns in SQL queries without warning about SQL injection risks. While the skill itself doesn't execute queries, it teaches unsafe SQL practices that could lead to SQL injection vulnerabilities when users implement the suggested patterns.
SKILL.md
Examples show direct value insertion like "WHERE status = 'active'" without discussing parameterized queries or prepared statements. No mention of SQL injection prevention, input sanitization, or use of parameterized queries throughout the extensive documentation.What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 10 KB |
package.json | config | 588 B |
references/examples.md | doc | 12 KB |
references/optimization.md | doc | 10 KB |
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/sql-assistant
- Raw scan result: sql-assistant-scan.json
- Badge for your README:
- Request a re-scan