Back to skill
Skillv2.1.1
ClawScan security
Database Designer · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 10, 2026, 8:10 AM
- Verdict
- benign
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's requested resources and runtime instructions match a database-design toolkit; nothing in the manifest or SKILL.md suggests unrelated privileges or secret access, but the bundled Python scripts were not fully visible so a code-level review is recommended before use in production.
- Guidance
- This package appears coherent with its stated purpose and is likely safe to inspect and run locally. Before installing or running against production data: 1) Inspect the three Python scripts (index_optimizer.py, migration_generator.py, schema_analyzer.py) for any network calls (requests/urllib/sockets), subprocess.exec/ Popen usage, or code that reads arbitrary filesystem locations or environment variables. 2) Run the tools on sample files first and review generated SQL rather than executing it. 3) If you intend to run migrations against a live DB, ensure you have backups and run in a staging environment first. Because some source files were not shown in full here, I have medium confidence — a quick code review for data-exfiltration or remote execution patterns would raise confidence to high.
Review Dimensions
- Purpose & Capability
- okName and description (schema analysis, index optimization, migration generation) align with the included files (README, references, sample schemas, expected outputs) and the three utility scripts. No environment variables, binaries, or external services are declared that would be unexpected for this purpose.
- Instruction Scope
- okSKILL.md and README instruct the agent/user to analyze local schema files, query-pattern files, and produce reports and SQL migration plans. The instructions do not request reading unrelated system paths, scanning shell history, or contacting external endpoints. They describe generating SQL and reports rather than executing changes directly.
- Install Mechanism
- okNo install spec (instruction-only) and the README indicates simple local execution with Python 3.7+. No external downloads, package installs, or archive extraction are present in metadata. This is low-risk from an install perspective.
- Credentials
- okThe skill declares no required environment variables, no primary credential, and no config paths. That is proportionate to a local analysis/generation tool. Note: if you plan to have the scripts connect to a live database, additional credentials would be required—current manifest does not request them.
- Persistence & Privilege
- okThe skill is not always-enabled, is user-invocable, and allows model invocation (the platform default). It does not request permission to modify other skills or global agent configuration in the manifest.