v6.11.0 Self Update 2026-06-14 Current version
- ▸ Chequeo de versión remota al arrancar
- ▸ Diálogo accionable X→Y + comparación semver
- ▸ Upgrade garantizado con verificación
- ▸ Camino de divergencia con backup
How SpecBox Engine Works
Complete spec-driven development pipeline with multi-agent orchestration: specialized agents, MCP tools, skills and quality gates.
4 commands. Verified software. Auditable evidence.
You frame the problem with /discovery (ICP + JTBD). You describe what to build with /prd. The engine plans with /plan. Implements with /implement. Quality gates verify every step. Self-healing fixes errors automatically. You review the PR.
Real example: from idea to PR in 4 steps
How it works: 4 commands, verified software
$ /discovery "Restaurant booking system"
Framing before building...
┌─ ICP: restaurant owner, no tech team
└─ JTBD: fill tables without calls or no-shows
→ icp_jtbd.md ✓ (feeds the PRD)
$ /prd "Restaurant booking system"
Generating PRD...
┌─ US-001: Book table online
│ ├─ UC-001: Select date and time (4 ACs)
│ └─ UC-002: Cancel booking (2 ACs)
└─ Quality Gate: PASS (12/12 ACs are specific and measurable)
$ /plan
Analyzing PRD (4 US, 12 UC, 38 AC)...
┌─ Phase 1: Database schema + RLS
├─ Phase 2: API (12 endpoints)
├─ Phase 3: UI (Stitch designs)
└─ Phase 4: E2E Playwright
Estimated: ~6h with SpecBox. Without: 3-5 days.
$ /implement
[Orchestrator] Creating branch feat/US-001
[AG-03] Migrations: create_bookings_table ✓
[AG-01] Implementing UC-001 from Stitch design...
[AG-04] Tests: 24/24 passing, coverage 87%
[AG-08] Quality Gate: GO ✓
[AG-09] Acceptance: 12/12 AC ACCEPTED
→ PR #47 created, ready for review
The Pipeline
$ /discovery
Product framing before the PRD: identifies the ICP (ideal customer) and extracts the JTBD (jobs to be done). Generates doc/discovery/<feature>/icp_jtbd.md, the input that steers /prd. Available since v6.0.0.
$ /prd
Generates the Product Requirements Document with User Stories, Use Cases, and Acceptance Criteria. Definition Quality Gate validates each criterion is specific, measurable, and testable.
$ /plan
Analyzes the PRD, generates a technical plan with phases, UI components, and Stitch designs. VEG generates visual directives tailored to the audience.
$ /implement
Autopilot: creates branch, executes sequential phases, design-to-code, quality gates between phases, acceptance testing, and automatic PR.
Deep Dive — Everything inside
Specialized Agents
Each pipeline phase has agents with defined roles. The Orchestrator NEVER writes code — it only coordinates, delegates, and consolidates.
last known snapshot
Agent Skills
Auto-discoverable commands that activate when relevant. Each skill is a complete workflow.
· last known snapshot
Automation Tools
Unified MCP server. Backend-agnostic: works with Trello, Plane, or locally without external APIs.
Each tool is an atomic operation agents use to manage your project: create PRDs, run tests, move cards, verify quality, generate evidence.
— MCP tools · last known snapshot
·
…
·
…
·
…
·
…
·
…
·
…
·
…
·
…
VS Code Extension
Install SpecBox Engine from the Marketplace. Guided onboarding, auto-update and GitHub sign-in.
🧩
View on Marketplace EmbedBuild.specbox-engine
v— · last known snapshot
Quality Gates
Controls that BLOCK the pipeline when the work doesn't meet the bar. They are not repairs: you either pass them, or you don't move on.
📐
Definition Quality Gate
In /prd: validates each Acceptance Criteria for specificity, measurability and testability. If the average is below the threshold, it rejects the work item and proposes improvements.
🔍
Quality Audit · AG-08
In /implement: ISO/IEC 25010 audit (SQuaRE, 8 characteristics). Verdict GO / CONDITIONAL GO / NO-GO. Blocks the merge if NO-GO.
✅
Acceptance Gate · AG-09
In /implement: checks that every AC is implemented, tested and evidenced. Verdict ACCEPTED / CONDITIONAL / REJECTED per Use Case. Blocks the merge if REJECTED.
🧪
Blocking Lint + Tests
Pipeline hooks (pre-commit-lint, spec-guard, e2e-gate) that prevent commit or merge with dirty lint or failing tests.
Self-Healing
When a phase fails, the engine tries to repair itself before asking for help. Real budget: max ~8 auto-heals per implementation and 2 attempts per phase; if 2 consecutive phases fail, it stops.
1
Auto-fix
Level 1 — fixes lint, formatting and imports automatically (e.g. dart fix --apply, eslint --fix).
2
Patch
Level 2 — diagnoses and fixes the root cause: missing import, wrong type, dependency or unresolved path.
3
Rollback
Level 3 — partial rollback: git stash, marks the checkpoint as failed and retries the phase from a clean state.
4
Human Intervention
Level 4 — escalates to the developer with a diagnosis (error_report.md) when the phase keeps failing after retries.
Pipeline Integrity
Hook-level enforcement that makes it impossible to write code without an active UC.
"The embed-build incident (March 2026): an agent implemented 9 Use Cases without the pipeline, leaving Trello empty with zero traceability. That was the day HARD BLOCKS were born."
🛡️
spec-guard.sh blocks Write/Edit to src/ without active UC
🚫
commit-spec-guard.sh blocks commits to main/master
🎨
design-gate.sh blocks UI without prior Stitch designs
💀
Anti-main guard: FATAL ERROR if implementing on main
Multi-Backend
4 interchangeable backends with the same interface (25 methods). Bidirectional migration between them.
📋
Trello Boards with US/UC/AC cards
✈️
Plane Cloud or self-hosted (CE)
📁
FreeForm Local JSON, no external API
🐘
Native Postgres/Supabase, multi-tenant
SpecBox Cloud The ecosystem web panel: observability and management at cloud.specbox.build.
Infrastructure Services
Integrated patterns for 5 services: each with configuration guides, best practices, and pipeline integration.
⚡ Supabase
🐘 Neon
💳 Stripe
🔥 Firebase
🔄 n8n
Supported Stacks
💙 Flutter
⚛️ React
🐹 Go
🐍 Python
📜 Google Apps Script
Engine Evolution
Every version is real and in production. Here's what changed most recently.
Want SpecBox on your team?
Get in touch and we'll show you how to adopt it in your workflow.
jesus@embed.build