The package does not show an install-time attack, but explicit CLI use can modify Claude Code permissions and configuration. Under the agent integration policy, this is a guarded capability risk rather than malware.
Static reason
One or more suspicious static signals were detected.
Trigger
A user runs the setup command, or runs the server command before global setup exists.
Impact
This is an explicit-command agent configuration capability that can enable later execution of the package's MCP integration; no install-time or covert activation was found.
Mechanism
The CLI writes MCP and Claude settings configuration, including a command that selects the latest DiffPrism release.
Evidence
.mcp.json.claude/settings.json.claude/skills/review/SKILL.md