Connect Claude Code to a focused, non-destructive, closed-world OrgX status profile through native OAuth.
Critical failure: Tool definitions are not visible in source code. The verify-plugin.mjs file references 7 tools by name only (get_initiative_pulse, get_operator_chronicle, get_agent_status, get_morning_brief, orgx_inspect, orgx_recommend, orgx_search) but contains no schema definitions, parameter descriptions, or output documentation. Tool registration appears to happen remotely via the HTTP endpoint (https://mcp.useorgx.com/mcp?profile=claude-directory), making static analysis impossible. Per hard scoring rules: if tool definitions are inferred rather than directly visible, cap per-tool scores at 50; if schemas are not visible, schema score must be 0; if descriptions are missing, description score must be 0. All 7 tools fail these gates.
No tool schemas visible in source code. Tool definitions are registered remotely; static analysis cannot verify input parameters, types, or constraints.
No tool descriptions visible in source code. LLMs cannot determine when or why to select each tool.
Naming inconsistency: 'orgx_inspect', 'orgx_recommend', 'orgx_search' use underscore-prefixed namespace but lack verb-noun clarity. 'orgx_inspect' is vague, inspect what? 'orgx_recommend', recommend what? Compare to clear names like 'get_initiative_pulse'.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 26 | 2026-07-28+ | v2 |
No output schema documentation visible. Agents cannot plan downstream calls or extract required fields without knowing what each tool returns.
No error handling guidance visible. Tools provide no recovery hints, error classification, or actionable next steps for LLMs.