Code Examples - MCPCore

Every example below is a complete, working tool body. Copy it into the code editor, add the listed parameters and secrets, and save.

HTTP — GET with authentication

Fetch data from a REST API using a Bearer token stored as a secret. Parameters: repo (string, required) Secrets: GITHUB_TOKEN

const res = await sdk.http({
  method: "GET",
  url: `https://api.github.com/repos/${params.repo}`,
  headers: {
    Authorization: `Bearer ${env.GITHUB_TOKEN}`,
    "User-Agent": "MCPCore-Tool/1.0",
  },
});

if (res.status() !== 200) {
  throw new Error(`GitHub API error ${res.status()}: ${res.body()}`);
}

const data = JSON.parse(res.body());

return {
  name:     data.full_name,
  stars:    data.stargazers_count,
  forks:    data.forks_count,
  language: data.language,
  topics:   data.topics,
};

HTTP — POST with JSON body

Send a JSON payload to an external API. Parameters: to (string, required), subject (string, required), body (string, required) Secrets: SENDGRID_API_KEY

const res = await sdk.http({
  method: "POST",
  url: "https://api.sendgrid.com/v3/mail/send",
  headers: {
    Authorization: `Bearer ${env.SENDGRID_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    personalizations: [{ to: [{ email: params.to }] }],
    from: { email: "noreply@yourapp.com" },
    subject: params.subject,
    content: [{ type: "text/plain", value: params.body }],
  }),
});

if (res.status() >= 400) {
  throw new Error(`SendGrid error ${res.status()}: ${res.body()}`);
}

return { status: res.status(), message: "Email sent successfully" };

HTTP — Webhook relay

Receive data from the AI and forward it to a webhook endpoint. Parameters: event (string, required), payload (object, required) Secrets: WEBHOOK_URL, WEBHOOK_SECRET

const res = await sdk.http({
  method: "POST",
  url: env.WEBHOOK_URL,
  headers: {
    "Content-Type": "application/json",
    "X-Webhook-Secret": env.WEBHOOK_SECRET,
  },
  body: JSON.stringify({
    event: params.event,
    data: params.payload,
    timestamp: new Date().toISOString(),
  }),
});

return {
  status: res.status(),
  accepted: res.status() >= 200 && res.status() < 300,
};

Database — Select with filters

Query a PostgreSQL table with optional filters and return results. Parameters: userId (string, required), status (string, optional) Secrets: DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const where = { user_id: params.userId };
if (params.status) {
  where.status = params.status;
}

const orders = await db.select("orders", where);

return {
  count: orders.length,
  orders,
};

Database — Insert a record

Insert a new row and return the created record. Parameters: name (string, required), email (string, required), role (string, optional) Secrets: DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const record = await db.insert("users", {
  name:       params.name,
  email:      params.email,
  role:       params.role ?? "member",
  created_at: new Date().toISOString(),
});

console.log("Created user:", record.id);
return { id: record.id, message: "User created" };

Database — Raw SQL with pagination

Use a raw parameterised query for complex operations with pagination support. Parameters: category (string, required), page (number, optional), pageSize (number, optional) Secrets: DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const page     = params.page ?? 1;
const pageSize = params.pageSize ?? 20;
const offset   = (page - 1) * pageSize;

const result = await db.query(
  `SELECT p.id, p.name, p.price, p.stock,
          c.name as category_name
   FROM products p
   JOIN categories c ON c.id = p.category_id
   WHERE c.slug = $1
   ORDER BY p.created_at DESC
   LIMIT $2 OFFSET $3`,
  [params.category, pageSize, offset]
);

const countResult = await db.query(
  `SELECT COUNT(*) as total
   FROM products p
   JOIN categories c ON c.id = p.category_id
   WHERE c.slug = $1`,
  [params.category]
);

return {
  page,
  pageSize,
  total: parseInt(countResult.rows[0].total),
  products: result.rows,
};

Database — MySQL example

Connect to a MySQL database and run a query. Parameters: email (string, required) Secrets: MYSQL_HOST, MYSQL_PORT, MYSQL_DB, MYSQL_USER, MYSQL_PASSWORD

const db = await sdk.db({
  type: "mysql",
  connection: {
    host:     env.MYSQL_HOST,
    port:     env.MYSQL_PORT,
    database: env.MYSQL_DB,
    user:     env.MYSQL_USER,
    password: env.MYSQL_PASSWORD,
  },
});

const users = await db.select("users", { email: params.email });

if (users.length === 0) {
  return { found: false, message: "User not found" };
}

return { found: true, user: users[0] };

Lodash — Data transformation

Use sdk.lodash to group, sort, and summarise data. Parameters: userId (string, required) Secrets: DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const orders = await db.select("orders", { user_id: params.userId });
const _ = sdk.lodash;

// Group orders by status
const grouped = _.groupBy(orders, "status");

// Get the 5 most recent orders
const recent = _.orderBy(orders, ["created_at"], ["desc"]).slice(0, 5);

// Calculate total spent
const totalSpent = _.sumBy(orders, "amount");

// Count orders by status
const statusCounts = _.countBy(orders, "status");

return {
  totalOrders: orders.length,
  totalSpent,
  statusCounts,
  recentOrders: recent,
};

Lodash — Deduplication and filtering

Clean and deduplicate data from an external API. Parameters: query (string, required) Secrets: API_KEY

const res = await sdk.http({
  method: "GET",
  url: `https://api.example.com/search?q=${encodeURIComponent(params.query)}`,
  headers: { Authorization: `Bearer ${env.API_KEY}` },
});

const data = JSON.parse(res.body());
const _ = sdk.lodash;

// Remove duplicates by ID
const unique = _.uniqBy(data.results, "id");

// Filter out inactive items
const active = _.filter(unique, item => item.status === "active");

// Pick only the fields we need
const cleaned = active.map(item => _.pick(item, ["id", "name", "category", "score"]));

// Sort by score descending and take top 10
const top10 = _.orderBy(cleaned, ["score"], ["desc"]).slice(0, 10);

return { count: top10.length, results: top10 };

Combined — API call + database write

Fetch data from an external API and store it in your database. Parameters: repo (string, required) Secrets: GITHUB_TOKEN, DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

// 1. Fetch from GitHub
const res = await sdk.http({
  method: "GET",
  url: `https://api.github.com/repos/${params.repo}`,
  headers: {
    Authorization: `Bearer ${env.GITHUB_TOKEN}`,
    "User-Agent": "MCPCore-Tool/1.0",
  },
});

if (res.status() !== 200) {
  throw new Error(`GitHub API error: ${res.status()}`);
}

const repo = JSON.parse(res.body());
console.log("Fetched repo:", repo.full_name);

// 2. Store in database
const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const record = await db.insert("repo_snapshots", {
  repo_name:  repo.full_name,
  stars:      repo.stargazers_count,
  forks:      repo.forks_count,
  language:   repo.language,
  captured_at: new Date().toISOString(),
});

console.log("Saved snapshot:", record.id);

return {
  repo: repo.full_name,
  stars: repo.stargazers_count,
  forks: repo.forks_count,
  snapshotId: record.id,
};

Error handling pattern

A robust pattern for handling multiple failure points. Parameters: userId (string, required) Secrets: API_URL, API_KEY, DB_HOST, DB_PORT, DB_NAME, DB_USER, DB_PASSWORD

// 1. Fetch user from external API
const res = await sdk.http({
  method: "GET",
  url: `${env.API_URL}/users/${params.userId}`,
  headers: { Authorization: `Bearer ${env.API_KEY}` },
});

if (res.status() === 404) {
  return { found: false, message: "User not found in external system" };
}

if (res.status() !== 200) {
  console.error("API error:", res.status(), res.body());
  throw new Error(`External API returned ${res.status()}`);
}

const user = JSON.parse(res.body());
console.log("Found user:", user.name);

// 2. Enrich with database data
const db = await sdk.db({
  type: "pg",
  connection: {
    host:     env.DB_HOST,
    port:     env.DB_PORT,
    database: env.DB_NAME,
    user:     env.DB_USER,
    password: env.DB_PASSWORD,
  },
});

const orders = await db.select("orders", { user_id: params.userId });
const _ = sdk.lodash;

return {
  found: true,
  user: {
    id:    user.id,
    name:  user.name,
    email: user.email,
  },
  stats: {
    totalOrders: orders.length,
    totalSpent:  _.sumBy(orders, "amount"),
    lastOrder:   _.maxBy(orders, "created_at")?.created_at ?? null,
  },
};

Documentation Index

​HTTP — GET with authentication

​HTTP — POST with JSON body

​HTTP — Webhook relay

​Database — Select with filters

​Database — Insert a record

​Database — Raw SQL with pagination

​Database — MySQL example

​Lodash — Data transformation

​Lodash — Deduplication and filtering

​Combined — API call + database write

​Error handling pattern