vibe-code-cleanup

Warn

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses various shell commands such as find, grep, rg, cat, npm, and npx to perform reconnaissance and cleanup on the project filesystem.
  • [DATA_EXFILTRATION]: The skill accesses sensitive application files containing credentials. Specifically, Step 5 instructs the agent to execute cat .env.local to audit environment variables, which exposes the contents of these secret files to the AI model's context.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection due to its automated reading of untrusted codebase files. Ingestion points: The agent reads source code and configuration files during several cleanup steps. Boundary markers: No delimiters or specialized instructions are used to distinguish file content from the agent's logic. Capability inventory: The skill has the ability to execute shell commands and modify the filesystem based on analyzed data. Sanitization: There is no process for validating or escaping content read from the files before processing.
  • [EXTERNAL_DOWNLOADS]: The skill fetches and runs common development tools such as typescript, ts-prune, and eslint via the npx command.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 16, 2026, 07:03 PM
Security Audit — agent-trust-hub — vibe-code-cleanup