skills/lovstudio/skills/lov-cc-mv/Gen Agent Trust Hub

lov-cc-mv

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches the migration utility @lovstudio/cc-mv from the official NPM registry.- [REMOTE_CODE_EXECUTION]: It uses npx -y to download and execute the migration tool at runtime, which bypasses package installation confirmation.- [COMMAND_EXECUTION]: The skill performs filesystem operations and executes shell commands to move project directories and rewrite Claude Code's internal state files (JSONL).- [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill reads session summaries using npx -y @lovstudio/cc-mv --list-sessions --json, which extracts firstUserPrompt content from existing Claude Code session files in ~/.claude/.- Boundary markers: There are no explicit instructions to use delimiters or ignore instructions embedded within the ingested session prompts.- Capability inventory: The skill has broad capabilities including shell command execution (npx), file movement, and modification of internal configuration files.- Sanitization: The skill does not implement sanitization for the retrieved session content before the agent evaluates it to match user migration requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:39 PM