XJTLUmedia/Context-First-MCP

**The MCP server that keeps your AI grounded, coherent, and honest — across every turn.**
*Works instantly with Claude Desktop · Cursor · VS Code · any MCP client · Vercel remote — zero API keys needed.*
---
> **37 research-backed tools** across 7 layers — context health, state, sandboxing, persistent memory, advanced reasoning, truthfulness verification, orchestration, structured research, and autonomous file export. One context_loop call replaces 6–7 individual tools and returns a unified action directive.
---
Why Your AI Conversations Break Down
Long AI conversations fail in predictable ways. Context-First fixes all four:
| Failure Mode | What Goes Wrong | Context-First Solution |
|---|---|---|
| **Context Drift** | AI forgets earlier decisions and intent as the conversation grows | context_loop + detect_drift continuously re-anchor every turn |
| **Silent Contradiction** | New inputs silently overrule established facts — the AI doesn't notice | detect_conflicts compares every input against locked ground truth |
| **Vague Execution** | AI proceeds on underspecified requirements, producing misaligned output | check_ambiguity + abstention_check ask clarifying questions instead of guessing |
| **Hallucinated Success** | Tool outputs *look* successful but didn't actually achieve the goal | verify_execution rechecks whether the outcome matches the stated intent |
---
What You Get
**37 production-ready tools** grouped into 7 layers — plus 1 orchestrator that runs them all:
**One call. One directive. One score.**
---
Quick Start
npx — zero install
Claude Desktop
Cursor / VS Code
Remote (Streamable HTTP)
Deploy your own Vercel instance
---
Tool Reference
Layer 1: Core Context Health (9 tools)
| Tool | Purpose |
|------|---------|
| context_loop | **One-call orchestrator.** Runs 8 stages (ingest→recap→conflict→ambiguity→entropy→abstention→discovery→synthesis) and returns a single directive with action, contextHealth score, extracted facts, and suggested next tools |
| recap_conversation | Extracts hidden intent, key decisions, and produces consolidated state summaries |
| detect_conflicts | Compares new input against ground truth; surfaces contradictions |
| check_ambiguity | Identifies underspecified requirements and generates clarifying questions |
| verify_execution | Validates whether tool outputs actually achieved the stated goal |
| entropy_monitor | Proxy-entropy scoring via lexical diversity, contradiction density, hedge frequency, and n-gram repetition (ERGO) |
| abstention_check | 5-dimension confidence scoring — abstains with questions rather than hallucinating (RLAAR) |
| detect_drift | Detects conversation drift from the original intent |
| check_depth | Evaluates response depth against question complexity |
Layer 1b: State Management (4 tools)
| Tool | Purpose |
|------|---------|
| get_state | Retrieve confirmed facts and task status |
| set_state | Lock in ground truth — subsequent conflict checks run against these values |
| clear_state | Reset specific keys or all state |
| get_history_summary | Compressed conversation history with intent annotations |
Layer 2: Sandbox & Discovery (3 tools)
| Tool | Method | Purpose |
|------|--------|---------|
| discover_tools | MCP-Zero + ScaleMCP | Natural-language tool routing — returns only semantically relevant tools, reducing context bloat by up to 98% |
| quarantine_context | Multi-Agent Quarantine | Create isolated memory silos for sub-tasks, preventing intent dilution |
| merge_quarantine | Multi-Agent Quarantine | Merge silo results with noise filtering — only promoted keys return to main context |
Layer 3: Persistent Memory (6 tools)
| Tool | Purpose |
|------|---------|
| memory_store | Store findings, decisions, and intermediate results with metadata |