ln-724-artifact-cleaner
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands like
rm -rfandrm -fto delete detected platform artifacts, such as.local/,.cache/, and.replitconfiguration files. - [EXTERNAL_DOWNLOADS]: Fetches reference files and shared configurations from the author's GitHub repository (
https://raw.githubusercontent.com/levnikolaevich/claude-code-skills/...) as a fallback if the required files are not found in the local skill directory. - [PROMPT_INJECTION]: The skill processes untrusted data from the project being cleaned (source code, configuration files), which presents a surface for indirect prompt injection.
- Ingestion points: Reads local project files including
package.json, build configurations (vite.config.ts), and source code files. - Boundary markers: No specific delimiters are used to separate user data from agent instructions during processing.
- Capability inventory: The skill has the ability to delete files, modify source code, and execute shell commands.
- Sanitization: Relies on specific glob patterns and regex matching to identify and target only known platform artifacts.
Audit Metadata