MCP Servers / memory / Contextful

Contextful

JSON →
stdionone2memory

Highly efficient context management for agentic AI: MCP code search, evidence packs, graph context, and memory for large projects.

Install
How to run this server
[ { "cmd": "npx -y @inferensys/contextful", "imports": [] } ]
server path: contextful
Tools
What this server exposes
context_pack
Returns a ranked, cited, token-budgeted bundle of context for a query, scope, and budget.
search_code
Powerful code, docs, symbol, and memory search with query, mode, and filters.
trace_path
Graph traversal across files, symbols, modules, and config from one node to another.
impact_analysis
Reverse dependencies and likely tests for a symbol or file.
why_changed
Current evidence plus git history for a symbol or file.
recall_memory
Search session learnings and durable project lessons by query and scope.
write_lesson
Store an evidence-backed memory with claim, evidence refs, and scope.
Configuration
Environment & auth
authnone
Resources
Contextful — MCP Server · libregistry