Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.
Static reason
No blocking static signals were detected.
Trigger
User runs `scribe mcp connect` or `scribe skills install`.
Impact
A user-selected MCP URL may be registered at Claude user scope; bundled instruction skills may be installed.
Mechanism
User-invoked Claude MCP registration and skill copying.
Rationale
No concrete malicious behavior was found after source inspection. The explicit user-scope agent integration is a real dangerous capability under the stated policy, so it warrants a warning rather than a block.
Evidence
package.jsonsrc/index.tssrc/commands/mcp.tssrc/commands/skills.tssrc/update-check.tssrc/commands/sync.ts~/.claude.json~/.claude/skills
Network endpoints2
localhost:3748/api/mcpregistry.npmjs.org/@scribe.cool/cli