Skip to main content

Pipeline skills reference

Pipeline skills execute sequentially to transform requirements into running code. Each skill produces artifacts consumed by the next stage.

Pipeline flow

requirements-engineer → specifications-engineer → spec-auditor →
plan-architect → task-generator → task-implementer → test-planner

requirements-engineer

Professional requirements engineering based on SWEBOK v4 Chapter 1.

Invocation

/sdd:requirements-engineer                    # Elicit mode (interactive)
/sdd:requirements-engineer --audit            # Audit mode (review existing)
/sdd:requirements-engineer --format           # Format mode (convert format)

Modes

mode
enum
default:"elicit"
Operation modeOptions:
  • elicit — Interactive requirements gathering
  • audit — Quality assessment of existing requirements
  • format — Convert between requirement formats

Input

input
string
Path to existing requirements document (audit/format modes only)

Output artifacts

  • requirements/REQUIREMENTS.md — Structured requirements in EARS format
  • Sections: Functional, Nonfunctional, Constraints, Traceability

EARS syntax

All requirements use Easy Approach to Requirements Syntax:
Ubiquitous
pattern
THE <system> SHALL <behavior>Example: THE system SHALL store all data encrypted at rest
Event-driven
pattern
WHEN <trigger> THE <system> SHALL <behavior>Example: WHEN a user submits login credentials THE system SHALL validate them within 2 seconds
State-driven
pattern
WHILE <state> THE <system> SHALL <behavior>Example: WHILE the system is in maintenance mode THE system SHALL reject all write operations
Unwanted
pattern
IF <condition> THEN THE <system> SHALL <behavior>Example: IF the database connection fails THEN THE system SHALL retry 3 times with exponential backoff

Quality gates

Each requirement is validated against:
  • Unambiguous — Single interpretation only
  • Testable — Concrete acceptance criteria
  • Atomic — Single decision per requirement
  • Binding — Stakeholder-confirmed necessity
  • True — Represents actual need, not solution

specifications-engineer

Transforms requirements into formal technical specifications.

Invocation

/sdd:specifications-engineer                          # Full generation
/sdd:specifications-engineer --analyze                # Readiness analysis
/sdd:specifications-engineer --brownfield             # From existing code

Modes

mode
enum
default:"create"
Operation modeOptions:
  • analyze — Analyze requirements for specification readiness
  • create — Generate specifications from requirements
  • validate — Validate existing specifications
  • brownfield — Incremental specification for existing code

Input

--requirements
string
default:"requirements/REQUIREMENTS.md"
Path to requirements document

Output artifacts

Generates complete spec/ directory:
spec/domain/
directory
Domain model and business rulesFiles:
  • 01-GLOSSARY.md — Ubiquitous language
  • 02-ENTITIES.md — Domain entities
  • 03-VALUE-OBJECTS.md — Value objects and enums
  • 04-STATES.md — State machines
  • 05-INVARIANTS.md — Business invariants (INV-XXX-NNN)
spec/use-cases/
directory
Use case specificationsFormat: UC-NNN-{slug}.md per use caseSections: Actor, Preconditions, Main Flow, Alternative Flows, Exception Flows, Postconditions
spec/workflows/
directory
Multi-step process flowsFormat: WF-NNN-{slug}.md per workflow
spec/contracts/
directory
API and event contractsFiles:
  • API-{module}.md — REST/GraphQL contracts per module
  • EVENTS-{module}.md — Domain events
  • PERMISSIONS-MATRIX.md — RBAC matrix
spec/tests/
directory
Test specificationsFiles:
  • BDD-UC-NNN.md — BDD scenarios per use case
  • PROPERTY-TESTS.md — Property-based test specs
spec/nfr/
directory
Non-functional requirementsFiles:
  • PERFORMANCE.md — Performance targets (p99, throughput)
  • LIMITS.md — Rate limits, quotas, thresholds
  • SECURITY.md — Security requirements
  • OBSERVABILITY.md — Logging, metrics, alerting
spec/adr/
directory
Architecture Decision RecordsFormat: ADR-NNN-{slug}.md per decision

Clarification markers

When ambiguities cannot be resolved during generation:
<!-- [NEEDS CLARIFICATION] NC-001: Max file size: 10MB or 25MB? -->
Tracked in spec/CLARIFICATIONS-PENDING.md until resolved.

spec-auditor

Systematic specification quality assurance and defect detection.

Invocation

/sdd:spec-auditor                                     # Full audit
/sdd:spec-auditor --fix                               # Apply corrections
/sdd:spec-auditor --focused --scope=CHANGE-REPORT.md  # Focused audit

Modes

mode
enum
default:"audit"
Operation modeOptions:
  • audit — Detect specification defects
  • fix — Apply corrections from answered audit
  • focused — Audit only changed documents

Defect categories

CAT-01
category
Ambiguities — Terms admitting multiple interpretationsSignals: “appropriate”, “reasonable”, “normally”, missing quantifiers
CAT-02
category
Implicit rules — Behavior given as understood but not documentedSignals: Validations not specified, order assumed
CAT-03
category
Dangerous silences — Cases not covered causing undefined behaviorSignals: Flows without error handling, states without exit transitions
CAT-04
category
Semantic ambiguities — Same term with different meaningsSignals: Synonyms not controlled, glossary violations
CAT-05
category
Contradictions — Specifications conflicting between documentsSignals: Different values for same parameter, incompatible flows
CAT-06
category
Incomplete specifications — Missing documents or placeholder sectionsSignals: TODOs, TBD sections, empty sections
CAT-07
category
Weak invariants — Critical rules without formal invariantSignals: Restrictions in text without INV-ID, invariants without validation
CAT-08
category
Evolution risks — Designs hindering future changesSignals: Hardcoded values, closed enums, missing versioning
CAT-09
category
Missing ADRs — Architectural decisions without documentationSignals: Technologies mentioned without justification

3C verification protocol

Spec-level verification dimensions:
Completeness
dimension
Spec coverage checks
  • Every REQ traces to at least one spec
  • No orphan specs without REQ
  • All spec subdirectories populated
  • No placeholder sections remain
  • Traceability chain intact
Correctness
dimension
Spec-requirement alignment
  • Semantic match with REQ intent
  • No contradictions between specs
  • All INV codes valid
  • State transitions consistent
  • Permission alignment
Coherence
dimension
Cross-spec consistency
  • Glossary adherence
  • Terminology uniformity
  • Valid cross-references
  • Value consistency
  • Format consistency

Output artifacts

audits/AUDIT-BASELINE.md
file
Baseline of accepted/deferred findings
audits/AUDIT-{version}.md
file
Audit report with findings by category and severity
audits/CORRECTIONS-PLAN-{version}.md
file
(Fix mode) Corrections plan with before/after

plan-architect

Generates FASE files and implementation plans from specifications.

Invocation

/sdd:plan-architect                                   # Full generation
/sdd:plan-architect --fase 0                          # Single FASE
/sdd:plan-architect --skip-clarify                    # Skip interactive clarification
/sdd:plan-architect --regenerate-fases                # Force FASE regeneration
/sdd:plan-architect --regenerate-fases --affected=1,5 # Cascade mode

Modes

mode
enum
default:"global"
Operation scopeOptions:
  • global — Generate all plan artifacts
  • per-fase — Generate plan for single FASE
  • skip-clarify — Use existing clarifications
  • regenerate-fases — Force FASE regeneration
  • audit-fases — Read-only coverage check

FASE files

FASE (implementation phase) files are navigation indices mapping specs to incremental phases.
plan/fases/FASE-N-{slug}.md
file
Implementation phase indexSections:
  • Objetivo — What the phase enables
  • Criterios de Éxito — Acceptance checklist
  • Specs a Leer — Organized by type (UCs, ADRs, Domain, Contracts)
  • Invariantes Aplicables — Relevant invariants
  • Contratos Resultantes — Endpoints and events to implement
  • Verificación — Validation commands
  • Alcance — Incluye/Excluye boundary table

Clarification taxonomy

10 categories of implementation gaps requiring clarification:
CL-TECH
category
Technology stack — Missing language/framework choices
CL-DATA
category
Physical data model — Logical-to-physical mapping gaps
CL-ARCH
category
Architecture topology — Undefined deployment strategy
CL-SEC
category
Security implementation — Specs without library/pattern
CL-INTEG
category
Integration patterns — Undefined external protocols
CL-PERF
category
Performance strategy — NFR targets without strategy
CL-TEST
category
Test implementation — Missing test framework
CL-CICD
category
Build & deploy — Missing CI/CD definition
CL-OBS
category
Observability — Missing monitoring/logging
CL-COST
category
Cost & resources — Missing infrastructure cost estimate

Output artifacts

plan/fases/
directory
FASE navigation indicesFiles:
  • README.md — Coverage matrices and dependency graph
  • FASE-N-{slug}.md — One per implementation phase
plan/PLAN.md
file
Master implementation planSections:
  • Technical Context
  • Component Decomposition
  • Cross-FASE Concerns
  • Risk Assessment
  • Developer Quickstart
  • Validation & Traceability
plan/CLARIFY-LOG.md
file
Interactive clarification session log
plan/ARCHITECTURE.md
file
Architecture viewsViews:
  • C4 System Context (L1)
  • C4 Container Diagram (L2)
  • C4 Component Diagram (L3)
  • Deployment View
  • Physical Data Model
  • Integration Map
  • Security Architecture
plan/fase-plans/PLAN-FASE-N.md
file
Per-FASE implementation detailsSections:
  • FASE-specific technical decisions
  • Component implementation details
  • API implementation notes
  • Data changes (migrations)
  • Test strategy
  • Test Coverage Map (source → test mapping)

task-generator

Decomposes implementation plans into atomic, reversible tasks.

Invocation

/sdd:task-generator                                   # Full generation
/sdd:task-generator --fase 0                          # Single FASE
/sdd:task-generator --regen                           # Regenerate from scratch
/sdd:task-generator --audit                           # Validate existing tasks
/sdd:task-generator --fase=1 --incremental            # Cascade mode

Task ID convention

format
string
TASK-F{N}-{SEQ}Examples:
  • TASK-F0-001 — First task of FASE 0
  • TASK-F3-015 — 15th task of FASE 3

Task markers

[P]
marker
Parallelizable — Can execute concurrently with other [P] tasks
blocks:
annotation
Tasks that depend on this task completing
blocked-by:
annotation
Tasks that must complete before this task
[RETROACTIVE]
marker
Task documents already-implemented code (from reverse-engineer)
Source: CASCADE-{id}
annotation
Task generated by req-change cascade

Task structure

Each task entry includes:
Commit
field
Pre-generated conventional commit messageFormat: {type}({scope}): {description}Types: feat, fix, refactor, test, chore, docs, ci, perf
Acceptance
field
Concrete acceptance criteria from spec
Refs
field
Traceability references (FASE, UC, ADR, INV, REQ)
Revert
field
Revert strategy and impact assessmentCategories: SAFE, COUPLED, MIGRATION, CONFIG
Review
field
Review checklist for human reviewers

Output artifacts

task/TASK-INDEX.md
file
Global index of all tasks across all FASEs
task/TASK-ORDER.md
file
Implementation order with dependency graph and critical path
task/TASK-FASE-N.md
file
Tasks for FASE N organized by phase:
  1. Setup — Project structure, dependencies
  2. Foundation — Shared infrastructure
  3. Domain — Entities, value objects, services
  4. Contracts — API routes, handlers, events
  5. Integration — Wiring, event handlers
  6. Tests — Unit, integration, BDD scenarios
  7. Verification — End-to-end validation

task-implementer

Implements code from task documents with test-first development.

Invocation

/sdd:task-implementer --fase 0                        # Implement FASE 0
/sdd:task-implementer --task TASK-F0-005              # Single task
/sdd:task-implementer --continue                      # Resume from last
/sdd:task-implementer --verify --fase 0               # Read-only verification
/sdd:task-implementer --fase 1 --new-tasks-only       # Cascade mode

Execution phases

Phase 0
phase
Context loadingLoads: CLAUDE.md, glossary, entities, states, invariants, FASE file, plan, task document
Phase 1
phase
Readiness gatesValidates: task document exists, plan exists, FASE exists, glossary exists, specs exist, git clean, tools available
Phase 2
phase
Task selection & orderingDetermines execution order respecting dependencies and [P] markers
Phase 3
phase
Pre-implementation designReads specs, extracts contracts, identifies invariants, analyzes integration
Phase 4
phase
Test-first constructionWrites failing tests for acceptance criteria, exception flows, and invariants (RED phase)
Phase 5
phase
ImplementationWrites code to pass tests (GREEN phase), then refactors (REFACTOR phase)
Phase 6
phase
Quality verificationExecutes review checklist, runs full test suite, verifies build and lint
Phase 7
phase
Atomic commit & SHA captureStages task files, commits with pre-generated message, captures SHA for traceability
Phase 8
phase
Progress trackingUpdates task checkbox, reports progress with SHA, evaluates next task
Phase 9
phase
FASE checkpoint & verificationCreates git tag, verifies Criterios de Éxito, runs full test suite, checks coverage

Pause conditions

Implementation pauses and asks user when:
PAUSE: Ambiguity
condition
Spec does not define behavior clearly
PAUSE: Decision Pending
condition
[DECISION PENDIENTE] marker found in spec
PAUSE: Spec-Implementation Conflict
condition
Spec and plan contradict
PAUSE: Build Failure
condition
Build fails after task implementation
PAUSE: Test Regression
condition
Previously passing test now fails

Implementation feedback

When spec-level issues are discovered during implementation:
feedback/IMPL-FEEDBACK-FASE-N.md
file
Spec-level issues found during implementationIssue types: AMBIGUITY, CONFLICT, MISSING-BEHAVIOR, INCORRECT-CONTRACT, STALE-DECISIONSeverities: BLOCKER, WARNINGFed to req-change for spec corrections

Output artifacts

src/**/*
directory
Implementation code following project structure
tests/**/*
directory
Unit tests, integration tests, BDD scenarios
task/TASK-FASE-N.md
file
Updated with completed task checkboxes (- [x])
feedback/IMPL-FEEDBACK-FASE-N.md
file
(if spec issues found) Spec-level feedback for req-change
Git tags
artifact
Checkpoint tags: fase-N-setup, fase-N-domain, fase-N-verified

test-planner

Generates comprehensive test plans and coverage maps.

Invocation

/sdd:test-planner                                     # Full test plan generation
/sdd:test-planner --audit                             # Gap detection only
/sdd:test-planner --fase 0                            # Single FASE scope

Output artifacts

test/TEST-PLAN.md
file
Master test planSections:
  • Test strategy by type (unit, integration, e2e, performance, security)
  • Coverage targets per FASE
  • Test environment requirements
  • Test data management
  • CI/CD integration
test/TEST-MATRIX-{domain}.md
file
Test coverage matrix per domain areaColumns: Requirement, Use Case, Test Type, Test File, Coverage Status
test/PERF-SCENARIOS.md
file
(if applicable) Performance test scenarios with load profiles and acceptance criteria

Build docs developers (and LLMs) love