TerminalSkills/skills

salesforce

Build integrations and automations with Salesforce.

View source

How to use

  1. Copy the installation command and review the source package before installing it in your own agent environment.
  2. Check the source document for required tools, credentials and supported hosts.
  3. Describe your input and desired deliverable; review the output against the source material.
npx skills add https://github.com/TerminalSkills/skills --skill salesforce

Before you start

  • Installation does not include a model subscription, API allowance or permission to access your files.
  • Repository stars and installation counts describe activity, not an independent quality or security test.
Original skill document

Rendered from the source repository. Headings, examples, code, tables, links, and referenced images are preserved.

Salesforce

Overview

Build on the Salesforce platform — the world's largest CRM. This skill covers REST and Bulk API integration, SOQL/SOSL queries, Apex triggers and classes, Flow automation, Platform Events for real-time messaging, Connected Apps for OAuth2, Salesforce DX for source-driven development, and deployment pipelines. Suitable for CRM integrations, custom business logic, and extending Salesforce with external systems.

Instructions

Step 1: Authentication — Connected App & OAuth2

Create a Connected App (Setup → App Manager → New Connected App), enable OAuth, set scopes: api and refresh_token (add web only for browser flows). Avoid full, which grants everything the user can do. Server-to-server use needs the JWT flow's certificate uploaded to the app and the app's users pre-authorized. Newer orgs steer new integrations to External Client Apps; the same flows apply there.

JWT Bearer Flow (server-to-server). Use https://test.salesforce.com as aud and token host for sandboxes, or your My Domain URL:

typescript
import jwt from "jsonwebtoken";
import fs from "fs";

async function getSalesforceToken(clientId: string, username: string) {
  const privateKey = fs.readFileSync("server.key", "utf-8");
  const token = jwt.sign({
    iss: clientId, sub: username, aud: "https://login.salesforce.com",
    exp: Math.floor(Date.now() / 1000) + 300,
  }, privateKey, { algorithm: "RS256" });

  const res = await fetch("https://login.salesforce.com/services/oauth2/token", {
    method: "POST", headers: { "Content-Type": "application/x-www-form-urlencoded" },
    body: new URLSearchParams({ grant_type: "urn:ietf:params:oauth:grant-type:jwt-bearer", assertion: token }),
  });
  const data = await res.json();
  return { accessToken: data.access_token, instanceUrl: data.instance_url };
}

API helper. Pick the version from the highest entry returned by GET {instanceUrl}/services/data/ and keep it in one place:

typescript
const API_VERSION = process.env.SF_API_VERSION ?? "60.0";

async function sf(method: string, path: string, token: string, instanceUrl: string, body?: any) {
  const res = await fetch(`${instanceUrl}/services/data/v${API_VERSION}${path}`, {
    method,
    headers: { Authorization: `Bearer ${token}`, "Content-Type": "application/json" },
    body: body ? JSON.stringify(body) : undefined,
  });
  if (!res.ok) throw new Error(`SF ${res.status}: ${await res.text()}`);
  return res.status === 204 ? null : res.json();
}

Step 2: SOQL & SOSL Queries

typescript
// Query with relationships
const opps = await sf("GET", `/query?q=${encodeURIComponent(`
  SELECT Id, Name, Amount, StageName, CloseDate, Account.Name, Owner.Name,
    (SELECT Id, Subject, Status FROM Tasks WHERE Status != 'Completed')
  FROM Opportunity WHERE StageName NOT IN ('Closed Won','Closed Lost') AND Amount > 50000
  ORDER BY CloseDate ASC`)}`, token, instanceUrl);

// Aggregate
const summary = await sf("GET", `/query?q=${encodeURIComponent(`
  SELECT StageName, COUNT(Id) cnt, SUM(Amount) total FROM Opportunity
  WHERE CloseDate = THIS_FISCAL_YEAR GROUP BY StageName`)}`, token, instanceUrl);

// Paginate large results
let url = `/query?q=${encodeURIComponent("SELECT Id, Name FROM Contact")}`;
const all = [];
while (url) {
  const data = await sf("GET", url, token, instanceUrl);
  all.push(...data.records);
  url = data.nextRecordsUrl || null;
}

// SOSL full-text search
const search = await sf("GET",
  `/search?q=${encodeURIComponent("FIND {TechStart} RETURNING Lead(Name,Email),Contact(Name),Account(Name)")}`,
  token, instanceUrl);

Step 3: REST API — CRUD & Composite

typescript
// Create Account + Contact + Opportunity
const account = await sf("POST", "/sobjects/Account", token, instanceUrl, {
  Name: "TechStart GmbH", Industry: "Technology", BillingCity: "Berlin", BillingCountry: "Germany",
});
await sf("POST", "/sobjects/Contact", token, instanceUrl, {
  FirstName: "Marta", LastName: "Schmidt", Email: "marta@techstart.io", AccountId: account.id, Title: "CTO",
});
await sf("POST", "/sobjects/Opportunity", token, instanceUrl, {
  Name: "TechStart — Enterprise License", AccountId: account.id,
  StageName: "Qualification", Amount: 150000, CloseDate: "2026-12-31",
});

// Update and upsert
await sf("PATCH", `/sobjects/Opportunity/${oppId}`, token, instanceUrl, { StageName: "Negotiation/Review", Amount: 175000 });
await sf("PATCH", `/sobjects/Account/External_ID__c/TECHSTART-001`, token, instanceUrl, { Name: "TechStart GmbH" });

// Composite API (up to 25 requests in one call)
await sf("POST", "/composite", token, instanceUrl, {
  allOrNone: true,
  compositeRequest: [
    { method: "POST", url: `/services/data/v${API_VERSION}/sobjects/Account`, referenceId: "newAccount", body: { Name: "NewCo" } },
    { method: "POST", url: `/services/data/v${API_VERSION}/sobjects/Contact`, referenceId: "newContact",
      body: { LastName: "Okafor", AccountId: "@{newAccount.id}" } },
  ],
});

Step 4: Bulk API 2.0

An upsert needs an external ID field (a custom field marked External ID); standard Email does not qualify.

typescript
const job = await sf("POST", "/jobs/ingest", token, instanceUrl, {
  object: "Contact", operation: "upsert", externalIdFieldName: "External_ID__c", contentType: "CSV",
});
await fetch(`${instanceUrl}/services/data/v${API_VERSION}/jobs/ingest/${job.id}/batches`, {
  method: "PUT",
  headers: { Authorization: `Bearer ${token}`, "Content-Type": "text/csv" },
  body: `FirstName,LastName,External_ID__c,AccountId\nMarta,Schmidt,TECHSTART-C-001,${accountId}`,
});
await sf("PATCH", `/jobs/ingest/${job.id}`, token, instanceUrl, { state: "UploadComplete" });

Step 5: Apex & Platform Events

Trigger (auto-task on high-value opportunities):

apex
trigger OpportunityTrigger on Opportunity (after insert) {
    List<Task> tasks = new List<Task>();
    for (Opportunity opp : Trigger.new) {
        if (opp.Amount >= 100000) {
            tasks.add(new Task(Subject='Review: ' + opp.Name, WhatId=opp.Id,
                OwnerId=opp.OwnerId, ActivityDate=Date.today().addDays(3), Priority='High'));
        }
    }
    if (!tasks.isEmpty()) insert tasks;
}

Custom REST endpoint:

apex
@RestResource(urlMapping='/custom/deals/*')
global class DealAPI {
    @HttpGet
    global static Map<String, Object> getActiveDealsSummary() {
        AggregateResult[] results = [SELECT StageName, COUNT(Id) cnt, SUM(Amount) total
            FROM Opportunity WHERE IsClosed = false GROUP BY StageName];
        List<Map<String, Object>> stages = new List<Map<String, Object>>();
        for (AggregateResult ar : results)
            stages.add(new Map<String, Object>{'stage'=>ar.get('StageName'),'count'=>ar.get('cnt'),'total'=>ar.get('total')});
        return new Map<String, Object>{'stages'=>stages, 'generatedAt'=>Datetime.now()};
    }
}

Always bulkify triggers (no SOQL or DML inside the loop) and keep one trigger per object. Platform Events — define Deal_Closed__e with fields in Setup, publish from Apex with EventBus.publish(event), subscribe externally via CometD Streaming API.

Step 6: SFDX & CI/CD

bash
sf project generate --name my-project && cd my-project
sf org login web --alias myorg
sf project retrieve start --source-dir force-app --target-org myorg
sf project deploy start --source-dir force-app --target-org myorg
sf apex test run --target-org myorg --test-level RunLocalTests --wait 10 --code-coverage --result-format human

GitHub Actions deploy:

yaml
steps:
  - uses: actions/checkout@v4
  - uses: actions/setup-node@v4
    with: { node-version: 22 }
  - run: npm install -g @salesforce/cli
  - run: |
      printf '%s' "$SF_AUTH_URL" > "$RUNNER_TEMP/auth.txt"
      sf org login sfdx-url --sfdx-url-file "$RUNNER_TEMP/auth.txt" --alias prod
      rm "$RUNNER_TEMP/auth.txt"
    env:
      SF_AUTH_URL: ${{ secrets.SF_AUTH_URL }}
  - run: sf project deploy start --source-dir force-app --target-org prod --test-level RunLocalTests

Step 7: Integration Patterns

Change Data Capture (SF → external). Enable the object under Setup → Change Data Capture. Subscribe with the Pub/Sub API (gRPC, current) or the CometD Streaming API; client below stands for a CometD client:

typescript
client.subscribe("/data/OpportunityChangeEvent", async (event: any) => {
  const { changeType, recordIds } = event.payload.ChangeEventHeader;
  if (changeType === "UPDATE" || changeType === "CREATE") {
    for (const id of recordIds) {
      const record = await sf("GET", `/sobjects/Opportunity/${id}`, token, instanceUrl);
      await externalDb.upsert("opportunities", { sf_id: id, name: record.Name, amount: record.Amount, stage: record.StageName });
    }
  }
});

External webhook → Salesforce (Stripe example):

typescript
// Verify the Stripe-Signature header on the raw body first (stripe.webhooks.constructEvent).
app.post("/webhook/stripe", async (req, res) => {
  if (req.body.type === "payment_intent.succeeded") {
    const payment = req.body.data.object;
    const opp = await sf("GET",
      `/query?q=${encodeURIComponent(`SELECT Id FROM Opportunity WHERE Stripe_Payment_ID__c = '${String(payment.id).replace(/['\\]/g, "\\$&")}'`)}`,
      token, instanceUrl);
    if (opp.records.length > 0) {
      await sf("PATCH", `/sobjects/Opportunity/${opp.records[0].Id}`, token, instanceUrl, { StageName: "Closed Won" });
    }
  }
  res.sendStatus(200);
});

Examples

Example 1: Lead-to-deal pipeline with automated follow-ups

User prompt: "Create an Apex trigger that auto-creates a follow-up task when a new opportunity over $100K is created, and write a SOQL query that gives me a pipeline summary grouped by stage with total amounts for this fiscal year."

The agent will write an after insert Apex trigger on Opportunity that checks Amount >= 100000 and creates a Task with 3-day due date, high priority, and the opportunity owner as assignee. It will also build a SOQL aggregate query using GROUP BY StageName with COUNT(Id) and SUM(Amount) filtered to THIS_FISCAL_YEAR, and show how to call it via the REST API to get the pipeline summary.

Example 2: Bi-directional sync between Salesforce and an external order system

User prompt: "Set up Change Data Capture to sync closed-won opportunities to our PostgreSQL database in real time, and create an inbound webhook endpoint that updates Salesforce opportunities when Stripe payments succeed."

The agent will subscribe to /data/OpportunityChangeEvent via CometD, filter for CREATE and UPDATE events, fetch the full record, and upsert it into PostgreSQL. For the inbound flow, it will create an Express endpoint at /webhook/stripe that extracts the payment ID from the Stripe event, queries Salesforce for the matching opportunity by custom field Stripe_Payment_ID__c, and patches the stage to "Closed Won".

Guidelines

  • Use the Composite API for related record creation — creating an Account, Contact, and Opportunity in a single composite call with @{referenceId.id} references is faster and ensures atomicity with allOrNone: true.
  • Paginate all SOQL queries — results over 2,000 records return a nextRecordsUrl; always loop until it is null to avoid silently missing data.
  • Prefer Bulk API 2.0 for large data loads — anything over a few hundred records should use the Bulk API with CSV upload to avoid governor limits and API call quotas.
  • Always test Apex with `RunLocalTests` — Salesforce requires 75% code coverage for production deploys; run sf apex test run --wait 10 (without --wait it returns a run ID at once) before every deployment to catch failures early.
  • Sanitize SOQL inputs — never interpolate user input directly into SOQL strings; use bind variables in Apex or parameterize queries to prevent SOQL injection.
  • Respect API limits — each org has a rolling 24-hour API request allocation; check GET /limits and batch with Composite or Bulk before polling in loops.
  • Keep secrets out of code — the JWT private key, consumer key and auth URL come from a secret store or environment variables, never the repository; use a dedicated integration user with a minimal permission set.
from this repository

More skills

All skills
TerminalSkills
Community

3dsmax-rendering

Scripts and configures production rendering in Autodesk 3ds Max with the V-Ray and Corona renderers: output size and files, render elements, denoising, light mix, batch and command-line rendering, and network rendering. Use when a user asks to set up a production render, render several cameras in one batch, render from the command line or on a render farm, add render passes for compositing, or cut render time for archviz and product shots.

installs
0
GitHub stars
180
Updated
Oct 25
TerminalSkills
Community

3dsmax-scripting

Covers scripting Autodesk 3ds Max, the 3D modeling and rendering application, with MAXScript and Python (pymxs): scene manipulation, object creation, material assignment, camera and light setup, batch operations, and file I/O. Use when tasks involve automating repetitive 3ds Max workflows, batch processing scenes, running scripts headless with 3dsmaxbatch, creating custom tools, or scripting scene setup for archviz, product visualization, or VFX.

installs
0
GitHub stars
180
Updated
Oct 25
TerminalSkills
Community

3proxy

3proxy is a small open-source proxy server that runs HTTP/HTTPS, SOCKS4/5, SNI and TCP/UDP port-mapping proxies from one config file. Use when a user asks to set up an HTTP or SOCKS5 proxy, add proxy users and passwords, write 3proxy access rules, chain or rotate upstream (parent) proxies, limit bandwidth, connections or monthly traffic per user, run 3proxy in Docker, or fix a 3proxy.cfg that will not start.

installs
0
GitHub stars
180
Updated
Oct 25
TerminalSkills
Community

a2a-protocol

Builds Agent2Agent (A2A) servers and clients, the open protocol (originally from Google, now under the Linux Foundation) that lets AI agents from different frameworks call each other. Use when the user wants to create an A2A-compliant agent, build an Agent Card, implement task management, connect agents across frameworks, set up agent discovery, handle streaming responses, implement push notifications, or orchestrate multi-agent workflows. Trigger words: a2a, agent to agent, agent2agent, a2a pro

installs
0
GitHub stars
180
Updated
Oct 25