Today4 stories · 1 brief

The Daily Org

Spring '27Sandboxes 6 Feb 2027

Curated by Abhinav (opens in a new tab)

Vol. I · No. 48The independent daily for Salesforce admins, developers and architects4 stories · 1 brief

Must-readAgentforce and AI

Six practices for taking Tableau MCP agents from prototype to production (opens in a new tab)

You will learn how to prepare a Tableau data model for agents, scope their tools and data, and build a test and monitoring harness before agents reach end users.

Why read this

The post draws on a regulated-industry customer deployment.

  1. Prepare the data model
  2. Define the agent's role
  3. Follow least privilege
  4. Load metadata into the prompt (optional)
  5. Pre-build queries for known questions
  6. Build evaluation and monitoring
Six practices for moving a Tableau MCP agent from prototype to production, in order

The post describes a large enterprise customer in a regulated industry that moved from Tableau dashboards to custom agents built on Tableau MCP. Off-the-shelf Tableau agents were too broad for the customer, who needed specialized workflows, agents that could act on results, embedding in a purpose-built application, and fine-grained control. Custom agents bring more need for testing, governance and monitoring, so the authors set out six practices.

The first is to prepare the data model. Business logic should live as close to the data as possible: calculated fields for fixed logic, Tableau Knowledge business preferences or field descriptions for logic shared across agents, and agent system instructions for logic specific to one agent. Teams should also remove duplicate fields, describe fields with units and caveats, name them in business terms, hide irrelevant fields and certify the data sources. The TDS Edit API, available in Tableau Cloud from October 2026 and Tableau Server 2026.3, lets teams add field descriptions in code.

The second is to define the agent’s role by answering what problem it solves, which domain it serves, what it can and cannot answer, and how it should respond. The answers shape its data, tools and system prompt. Tableau MCP now supports embedded data sources in 2026.3.

The third is to apply least privilege. The authors removed tools the agent did not need, starting with list-datasources, and pinned the agent to an approved data source ID. Tool scoping is set through environment variables on self-hosted deployments or site settings on hosted Tableau Cloud, and embedded data sources are scoped by project or tag.

The fourth, optional, practice is to load metadata into the system prompt at session start. This removed the get-datasource-metadata call and cut response time by almost 50 percent. The authors warn that this suits agents with one or two data sources and that stale metadata is a risk if schemas change often.

The fifth is to pre-build queries for frequent questions and expose them as custom tools, so the model only explains results. The authors advise against chasing perfect accuracy for open-ended questions, and suggest dashboards for figures that must be exact.

The sixth is an evaluation and monitoring framework: a question bank with correct answers, including refusals, run against each prompt, model and setting version, with LLM-as-a-judge grading for less deterministic answers, release-style versioning, production monitoring of latency, tokens and failures, and a regular review that adds new questions. The authors recommend an established agent observability platform, and they name LangSmith as an evaluation tool they used.

Read the original at Salesforce Blog (opens in a new tab)

Worth reading?

RecommendedAgentforce and AI

Deterministic gates let an AI agent approve production changes in seconds (opens in a new tab)

You will see how Salesforce's ROAM agent automates Change Advisory Board reviews by letting deterministic code decide while the LLM only explains.

Why read this

You will also learn the integration, contract-testing and fail-closed patterns that made it trustworthy.

  1. Verify approver identity and role
  2. Check linked passing QA test run
  3. Confirm pull request and approval trail
  4. Route to human if any gate is unverified
  5. Write every outcome to audit table
ROAM's gate verifies each requirement independently, and routes unverifiable requests to a human

Salesforce Engineering describes ROAM, the Release Operations Agentic Manager, an Agentforce-pattern agent that reviews production patches for 18 Salesforce products. Before automation, requesters and reviewers chased approvals across Slack, change records, QA evidence and release calendars, with median turnaround around 24 hours. The team mapped the Change Advisory Board process first, then built ROAM on an internal platform providing event ingestion, durable workflow state and shared clients.

The main technical lesson concerns integration correctness. An early bug read the wrong field in nested Slack replies and silently dropped approvals, so the model received an incomplete picture. To guard against schema drift, the team added contract tests using recorded live payloads that run on every pull request, and a mismatch fails the build. The core design rule is that the gate decides and the LLM narrates: deterministic code verifies the approver’s identity and role, the linked test run, the pull request and the approval trail, and the model cannot produce or overturn the verdict. Output validation re-reads the change record and fails closed when an explanation contradicts the evidence.

When a gate cannot be verified, for example because of missing links, expired tokens or outdated rosters, ROAM routes the request to a human, which happens for about 8% of requests. Idempotency keys and decision hashes prevent duplicate posts and changing verdicts on retries, and every outcome is written to a durable audit table.

On AI-assisted development, the article notes that roughly 70% of ROAM’s codebase was drafted or refined with Agentforce-assisted coding, while engineers wrote the remaining 30%, and a separate verifier agent re-executes test suites rather than trusting a fix agent’s self-report. Over about 550 reviews in 10 weeks, the team reports zero incorrect approvals, a median response of about 12 seconds and a 95th percentile under 45 seconds, after which ROAM became mandatory for all production patches.

Read the original at Salesforce Engineering (opens in a new tab)

Worth reading?

DevOps

Running Apex tests locally with Renmy Runtime, without an org (opens in a new tab)

You will learn why Apex has no local run loop, what a local Apex runtime changes for testing, debugging and CI, and which platform rules still apply locally.

Why read this

The article is a vendor-led introduction to a third-party tool, so weigh its claims accordingly.

The article explains that Apex looks like Java but only compiles, runs and tests inside a Salesforce org, so every iteration means deploying, waiting and reading logs. It argues that this deploy-to-run loop is slow, and that CI pipelines need an org and certificate just to run unit tests. Its benchmark claims four open-source Apex suites of 837 tests took about 1 min 46 s in an org versus 2.4 s locally, with every test passing in both.

The author presents Renmy Runtime, a single binary that executes Apex, SOQL, DML and Salesforce Flows on a developer’s machine against an embedded PostgreSQL, with no org, Docker or JVM. Commands such as renmy test, renmy validate and renmy explain run tests, type-check changes and explain failures, and the tool reports line and branch coverage. A JetBrains plugin adds breakpoints in IntelliJ IDEA, a SOQL and Apex console, and mutation testing as a Pro feature. The CI example runs the suite and emits JUnit XML and Cobertura without an org.

The piece is candid that some differences remain: governor limits, order of execution and metadata-defined schemas still apply, and the runtime is not the platform itself. It recommends keeping sandboxes or Scratch Orgs for UI, integration and final validation, with renmy deploy checking the same bytes in the org before release.

Read the original at dev.to (opens in a new tab)

Worth reading?

Flow and admin

Rolling out multi-factor authentication across a Salesforce org in stages (opens in a new tab)

You will know how to choose verification methods, phase an MFA rollout to avoid a backlash, handle integration and exception accounts, and prepare recovery procedures before users get locked out.

  1. Pick verification method
  2. Pilot with administrators
  3. Fix friction over one week
  4. Roll out profile by profile
  5. Review enrollment monthly
A staged MFA rollout runs from a pilot to profile-by-profile enablement

The article opens by framing MFA as a requirement for Salesforce sign-in, since a password alone no longer protects customer data in an org. It then sets out a five-part plan for the admin running the rollout. The first step is choosing a verification method: the Salesforce Authenticator app as the smooth default, a third-party TOTP app, or a physical security key, with keys recommended for System Administrators and users with broad data access.

The second step is staging the rollout. The author suggests enabling MFA under Setup, Identity Verification, starting with System Administrators and a small pilot group, running the pilot for a week, fixing friction such as missed enrollment emails or shared devices, and then rolling out profile by profile rather than all at once.

The third and fourth steps cover exceptions and non-human access. Users who cannot use a phone app, such as shared service accounts or kiosk logins, should use a security key or hardware token with a documented owner and review date. API-only and integration users should rely on OAuth flows or locked-down integration users, and the Login History report should be reviewed monthly for logins that bypassed MFA.

The final step is preparing users for recovery, since a new phone lockout is the most common support ticket. The author recommends a one-page guide and admin reset of MFA registration from the user’s record. The article closes by treating MFA as ongoing work, with monthly enrollment checks and quarterly exception reviews.

Read the original at dev.to (opens in a new tab)

Worth reading?

In brief