SDD Lifecycle

The "North Star" for your project. Before any feature work begins, the TPO executes a project-wide scan to establish architectural guardrails. This locks metadata conventions, deployment rules, and architectural posturing into a permanent memory state that the AI will respect forever.

Strategic Requirement Engineering. The TPO writes a functional specification that defines WHAT the feature must do from a business perspective. No technical logic is allowed here; the focus is entirely on user value and business logic.

The Reality Check. The Architect and TPO run a discovery cycle to see if the active Salesforce Org has any manual changes (Drift) that conflict with the specification. Gaps are identified and the Spec is refined before any technical planning starts.

Technical Architecture Blueprinting. The Architect generates the implementation strategy, defining exactly which classes, fields, and flows will be created or modified. This stage provides the AI with its technical execution sequence.

Atomic Decomposition. The technical plan is broken down into self-contained story files. These are Jira-ready units of work that can be distributed across a development team for parallel execution.

Human-in-the-Loop Validation. No automated code is written until a human confirms that the planned technical architecture perfectly serves the original functional specification. Stakeholder sign-off happens here.

Autonomous Building. The Developer executes the story files. The SFSpeckit engine enters an "Auto-Heal" cycle, writing code, running tests, fixing errors, and iterating until the technical plan is fulfilled.

Quality Evidence. The Developer generates a formal verification report. This includes PMD analysis, code coverage heatmaps, and performance metrics, serving as the "proof of quality" for the PR.

The Contribution Gate. Automated Pull Request preparation via GitHub CLI. The toolkit aggregates all implementation details and verification evidence into a standardized PR template.

Persona-Based Validation. The QA team executes multi-persona validation scripts. They ensure that the new feature works correctly for all specified user roles and that field-level security (FLS) is strictly enforced.

Business Value Sign-off. The Business Product Owner (BPO) or Stakeholders run automated UAT scripts generated by the tool to verify that the business objective has been achieved in the staging org.

Platform Sanity. A full suite of regression tests is executed against the feature branch to ensure that new code hasn't introduced breaking changes into existing production features.

Health Dashboard. Real-time project health metrics are aggregated. Quality scores across all stories are visualized to determine if the feature is ready for the promotion gate.

Promotion Gateway. Evidence-based release promotion across the org landscape (Staging -> Prod). Metadata is synced and dependencies are validated in the target environment.

Final Delivery. The toolkit automatically aggregates story summaries, functional intents, and testing results into a formal, business-ready Release Note document for stakeholders.