Test Harness Entry Diagnostics
PassAudited by ClawScan on May 5, 2026.
Overview
This instruction-only testing diagnostic is purpose-aligned, but it may read or edit code and run local build/test commands, so users should use it in a safe test environment.
This skill appears safe and coherent for legacy-code testing diagnostics. Before installing or using it, be aware that it may ask the agent to read your code, create or edit a small test, and run compile/test commands. Use a non-production workspace and review commands or edits before execution.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may inspect the codebase, create or modify a test, and run build/test commands locally.
The skill requests code-reading/search tools, shell access, and optional editing so it can create and run diagnostic tests. This matches the stated testing purpose, but users should notice the local command and edit authority.
tools-required: [Read, Grep, Bash]\n tools-optional: [Edit]
Use it on a test branch or disposable workspace, review any proposed edits or shell commands, and avoid running against production-connected configurations.
If the project is configured to use real services, a diagnostic construction test could trigger unintended local or external side effects.
The diagnostic intentionally runs constructor or method code to discover obstacles. In legacy systems, that can trigger side effects such as database connections or slow operations, which the skill itself acknowledges.
The runtime tells you exactly which side effects fire.
Run diagnostics with test doubles, local-only configuration, or a sandboxed test environment before allowing commands that instantiate legacy classes.