Connect ALOC examination content and AI-enriched assessment knowledge directly into Claude Desktop, Cursor, and autonomous agent frameworks.
The Model Context Protocol (MCP) is an open standard that enables Large Language Models (LLMs) and AI agents to securely discover and invoke tools, data sources, and APIs.
ALOC provides an official MCP Server available both as a hosted Streamable HTTP endpoint (https://aloc.com.ng/mcp/v1) and as a local npm package (@massteck/aloc-mcp-server) listed on Smithery.ai. It exposes structured tools for querying WAEC, JAMB, NECO, and Post-UTME examination content, step-by-step solutions, topic taxonomies, and Bloom cognitive difficulty ratings.
Tool discovery and cataloging consume 0 credits and require no API key. Metered tool executions consume standard credits (L1–L4) and authenticate securely via the X-API-Key or Authorization: Bearer header.
The server registers 12 specialized tools covering data retrieval, assessment generation, cognitive evaluation, and curriculum navigation:
aloc_get_platform_overviewDiscovery · Free (0 Credits)Self-describing overview of supported African examination bodies (WAEC, JAMB, NECO, Post-UTME), subject list, year coverage, and API rules.
{}aloc_get_sitemapDiscovery · Free (0 Credits)Returns direct URLs and deep links to developer documentation, error code references, and developer dashboard pages.
{}aloc_get_subject_yearsDiscovery · Free (0 Credits)List available examination years for a given subject and exam body.
{ subject: string, examType?: 'waec'|'jamb'|'neco'|'post-utme' }aloc_check_credit_balanceAccount · Free (0 Credits)Check developer key authentication status and remaining quota credits.
{}aloc_search_questionsL1 / L2 · 1–3 CreditsSearch curriculum-aligned examination questions by subject, examination body, year, topic, and difficulty score.
{ subject: string, examType?: 'waec'|'jamb'|'neco'|'post-utme', year?: number, topic?: string, difficulty?: 'easy'|'medium'|'hard', limit?: number }aloc_generate_assessmentL1 · 1 CreditAssemble a complete psychometrically balanced examination paper based on syllabus presets (JAMB 40, WAEC 50, Micro 10) or custom Bloom difficulty distributions, topics, and seeds.
{ subject: string, examType: string, preset?: 'jamb_standard_40'|'waec_standard_50'|'micro_test_10'|'custom', count?: number, seed?: string, shuffleOptions?: boolean, topics?: string[] }aloc_create_assessment_sessionL1 · 1 CreditInitialize a timed candidate assessment session (2-hour lifecycle) for cheat-resistant live test delivery.
{ subject: string, examType: string, preset?: string, count?: number, seed?: string, shuffleOptions?: boolean }aloc_get_session_questionsL1 · 1 CreditRetrieve questions allocated to an active candidate assessment session by session ID.
{ sessionId: string }aloc_get_question_explanationL3 · 10 CreditsRetrieve verified step-by-step explanations, common misconceptions, and remedial hints for a given question UUID.
{ questionId: string, depth?: 'step_by_step'|'concise'|'remedial' }aloc_get_curriculum_taxonomyL2 · Free / 3 CreditsExplore the official curriculum hierarchy, syllabus nodes, and topic classifications.
{ subject?: string, examType?: string }aloc_find_similar_questionsL4 · 10 CreditsRetrieve semantically and conceptually related variant questions using vector embeddings.
{ questionId: string, limit?: number }aloc_report_question_issueQuality · Free (0 Credits)Programmatically report typos, incorrect answer keys, or ambiguous phrasing to ALOC review pipeline.
{ questionId: string, type: 'question'|'option_a'|'correct_answer'|'explanation'|'other', message: string, reporterName?: string }The fastest way to configure your AI environment is through your developer dashboard:
Skip manually copying environment variables. The Developer Portal features a dedicated MCP Quick Connect card that generates your personalized JSON config with your active API key pre-filled.
Go to API Keys & MCP Quick Connect →Connect Claude Desktop using either the official Smithery.ai automated CLI, the hosted Streamable HTTP gateway, or local stdio:
npx -y @smithery/cli install massteck/aloc --client claudeAdd directly to your claude_desktop_config.json without downloading any npm packages locally:
{
"mcpServers": {
"aloc": {
"url": "https://aloc.com.ng/mcp/v1",
"headers": {
"X-API-Key": "YOUR_ALOC_API_KEY"
}
}
}
}{
"mcpServers": {
"aloc": {
"command": "npx",
"args": ["-y", "@massteck/aloc-mcp-server"],
"env": {
"ALOC_API_KEY": "YOUR_ALOC_API_KEY"
}
}
}
}Restart Claude Desktop. You will see all 12 ALOC tools appear under the hammer icon.
In Cursor, create a .cursor/mcp.json file in your repository root. You can connect using either the remote HTTP gateway or local stdio:
{
"mcpServers": {
"aloc": {
"url": "https://aloc.com.ng/mcp/v1",
"headers": {
"X-API-Key": "YOUR_ALOC_API_KEY"
}
}
}
}{
"mcpServers": {
"aloc": {
"command": "npx",
"args": ["-y", "@massteck/aloc-mcp-server"],
"env": {
"ALOC_API_KEY": "YOUR_ALOC_API_KEY",
"ALOC_AGENT_FRAMEWORK": "cursor"
}
}
}
}Drop a .cursorrules file into your project root to teach Cursor Composer and Claude how to leverage ALOC tools:
# .cursorrules — ALOC Assessment Infrastructure Assistant
You are an expert full-stack EdTech engineer building on ALOC Station (https://aloc.com.ng).
## Core Capabilities
- ALOC provides curriculum-aligned past and synthetic questions for WAEC, JAMB/UTME, NECO, and Post-UTME.
- You have access to @massteck/aloc-mcp-server with 12 tools for question search, assessment construction, and L3 step-by-step reasoning.
## Best Practices
1. Assessment Generation: Use 'aloc_generate_assessment' with official presets ('jamb_standard_40', 'waec_standard_50', 'micro_test_10') instead of raw single-question loops.
2. Anti-Cheating: Always pass 'shuffleOptions: true' and unique deterministic seeds ('seed') when creating candidate sitting mock tests.
3. Socratic Tutoring: When helping students, use 'aloc_get_question_explanation' to extract misconception hints rather than immediately revealing correct answer keys.
4. SDK Integration: In TypeScript projects, use '@massteck/aloc-sdk' (AssessmentsApi, QuestionsApi, ExplanationsApi).
Our MCP server implements standard MCP prompts to guide LLMs with pedagogical best practices:
Enacts an adaptive Socratic tutor for African curriculum exams (JAMB, WAEC, NECO). Queries questions via ALOC tools and provides progressive hints without spoiling answers.
Analyzes curriculum topic weightings and difficulty scores to compile a structured, multi-week study and mock testing timetable.
Evaluates candidate responses, diagnoses cognitive levels (Bloom's Taxonomy), and maps common distractors to underlying misconceptions.
For custom LLM pipelines (LangChain, LlamaIndex, or raw Node.js/Python MCP stdio clients), spawn the server directly:
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";
const transport = new StdioClientTransport({
command: "npx",
args: ["-y", "@massteck/aloc-mcp-server"],
env: { ALOC_API_KEY: process.env.ALOC_API_KEY },
});
const client = new Client({ name: "my-ai-tutor", version: "1.0.0" }, { capabilities: {} });
await client.connect(transport);