MCP server for Cortex Hub — real-time agent orchestration, code intelligence, memory management, and knowledge search via WebSocket conductor
18 tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Most tools have 1-3 parameters with type info, but descriptions are often generic (10-50 chars). No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Error handling is absent, no recovery guidance or actionable error messages. Tool names follow verb_noun convention (cortex_*), which is good, but parameter naming lacks consistency (e.g., 'repo' vs 'repository'). Output schemas are not documented in source. Composition is reasonable, tools are single-purpose, but chaining IDs are not always returned.
Get detailed information about code changes
Get contextual code information for a file or symbol
Analyze the impact of code changes on dependent code
Trigger re-indexing of a project after code changes. Looks up the project by repo URL and starts a GitNexus re-index job. Call this after pushing significant code changes to keep code intelligence up-to-date.
Search for code symbols and definitions across indexed repositories
Execute Cypher queries against the code knowledge graph
No output schemas documented. Tools return data but LLMs cannot plan downstream calls or extract fields without trial-and-error. E.g., cortex_memory_search, cortex_code_search, cortex_list_repos lack return type definitions.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint). LLMs cannot distinguish safe reads from destructive writes without explicit hints. cortex_memory_delete, cortex_session_end, cortex_code_reindex are marked DESTRUCTIVE/WRITE but lack annotation metadata.
Parameter descriptions are generic or missing context. E.g., cortex_cypher 'query' param says 'Cypher query to execute' but does not explain what graph it queries, what nodes/edges exist, or example patterns. cortex_code_reindex 'branch' param lacks format guidance (default: main).
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2026-07-28+ | v2 |
Detect code changes between two commits or branches
Check health status of Cortex Hub services
Search the Cortex knowledge base
Store knowledge in the Cortex knowledge base
List all indexed repositories in Cortex Hub
Delete a memory item from the Cortex memory system
Search for memory items in the Cortex memory system
Store a memory item in the Cortex memory system
Evaluate quality of a development plan
Generate a quality report for code changes
End a Cortex session
Start a new Cortex session for tracking agent work
No error handling or recovery guidance. Tools do not document what errors can occur, whether they are retryable, or what the LLM should do next. E.g., cortex_code_search could fail if repo is not indexed, but no guidance is provided.
Pagination not documented. cortex_memory_search, cortex_code_search, cortex_list_repos accept 'limit' but do not document offset/cursor, total count, or how to fetch next page. Large result sets risk context window exhaustion.