io.github.eidetic-works/nucleus
io.github.eidetic-works/nucleus · v1.17.1 · score 89/100 · checked 2026-11-22
Automated, read-only signal from public data, not an audit or an accusation. Maintainers can dispute any finding. How grading works
Persistent memory for AI agents — shared .brain across Claude Code, Cursor & Codex
Source: https://github.com/eidetic-works/nucleus-mcp
Findings
| Severity | Rule | Detail |
|---|---|---|
| Low | MW005 | Remote endpoint lists 28 tool(s) to anyone, no credentials needed. |
| Medium | MW102 | 1 tool(s) can execute commands or delete data (nucleus_plan_execute). Require confirmation before the agent calls them. |
What was checked
- https://relay.nucleusos.dev/mcp: answers without auth
- pypi: nucleus-mcp@1.17.1 (static checks only; code not executed)
Public tools: nucleus_governance, nucleus_features, nucleus_sessions, nucleus_tasks, nucleus_sync, nucleus_next_message, nucleus_relay_subscribe, nucleus_ccr_arm, nucleus_wakeup_wait, nucleus_orchestration, nucleus_telemetry, nucleus_slots, nucleus_infra, nucleus_agents, nucleus_federation, nucleus_engrams, nucleus_relay, nucleus_route, nucleus_audit, nucleus_delegate, nucleus_lane_init, nucleus_lane_start, nucleus_lane_stop, nucleus_lane_status, nucleus_lane_feedback, nucleus_plan_execute, nucleus_plan_import, nucleus_plan_list
Maintainer? Add the badge
Grades refresh daily. Think a finding is wrong? Dispute this grade.
[](https://www.mcpwatch.io/s/io-github-eidetic-works-nucleus-6ad3b08d)