Managing Spx

Foundational understanding required:

• /understanding-durable-map - Specs are permanent product documentation, not work items
• /understanding-outcome-decomposition - What belongs at each level (capability/feature/story)

If you find yourself wanting to "close" items or thinking in tasks, read these first.

• Template access → Read <accessing_templates> FIRST to understand where templates are located
• Structure definition → Read <structure_definition> for spx/ directory hierarchy, BSP numbering, test co-location
• ADR templates → Read <adr_templates> for Architectural Decision Record patterns
• PDR templates → Read <pdr_templates> for Product Decision Record patterns
• PRD templates → Read <requirement_templates> for Product Requirements
• Work item templates → Read <work_item_templates> for capability, feature, and story patterns

Use progressive disclosure - read only what you need.

When this skill is invoked, you are provided the base directory in the loading message:

Base directory for this skill: {skill_dir}

Use this exact path for all file access. Throughout this documentation, ${SKILL_DIR} is a placeholder—Claude must substitute it manually from the loading message.

IMPORTANT: Do NOT search the project directory for skill files.

All templates are under ${SKILL_DIR}/templates/:

${SKILL_DIR}/
├── SKILL.md                                    # This file
└── templates/
    ├── product/
    │   └── product.prd.md
    ├── decisions/
    │   ├── architectural-decision.adr.md
    │   └── product-decision.pdr.md
    └── outcomes/
        ├── capability-name.capability.md
        ├── feature-name.feature.md
        └── story-name.story.md

Always use the skill's base directory, not the user's project directory.

# Pattern
Read: ${SKILL_DIR}/templates/{category}/{template-name}

# Example: Read feature template
Read: ${SKILL_DIR}/templates/outcomes/feature-name.feature.md

1. Requirements (PRD) - Capture product vision without implementation constraints
2. Decisions (ADR/PDR) - Constrain architecture (ADR) or product behavior (PDR) with explicit trade-offs
3. Work Items (Capability/Feature/Story) - Sized, testable implementation containers

spx/
├── {product-name}.prd.md             # Product requirements
├── NN-{slug}.adr.md                  # Product-wide ADRs (interleaved)
├── NN-{slug}.pdr.md                  # Product-wide PDRs (interleaved)
└── NN-{slug}.capability/
    ├── {slug}.capability.md
    ├── tests/
    ├── NN-{slug}.adr.md              # Capability-scoped ADRs (interleaved)
    ├── NN-{slug}.pdr.md              # Capability-scoped PDRs (interleaved)
    └── NN-{slug}.feature/
        ├── {slug}.feature.md
        ├── tests/
        ├── NN-{slug}.adr.md          # Feature-scoped ADRs (interleaved)
        ├── NN-{slug}.pdr.md          # Feature-scoped PDRs (interleaved)
        └── NN-{slug}.story/
            ├── {slug}.story.md
            └── tests/

• Capability: Product-wide impact, contains features

  • Tests co-located in spx/NN-{slug}.capability/tests/
  • May have optional PRD as catalyst/enrichment

• Feature: Specific functionality, contains stories

  • Tests co-located in spx/.../NN-{slug}.feature/tests/
  • Technical details documented in feature.md (no separate TRD)

• Story: Atomic implementation unit, no children

  • Tests co-located in spx/.../NN-{slug}.story/tests/

Test level (unit, integration, e2e) is determined per-component by the evidence needed, not by hierarchy level. See /testing for the methodology.

• Requirements immutable - code adapts to requirements, not vice versa
• BSP numbering: Binary Space Partitioning encodes dependency order—lower BSP items are dependencies that higher-BSP items may rely on; same BSP means independent
• Full paths required: BSP numbers repeat at different levels (see <bsp_full_paths> below)
• Test co-location: Tests stay in spx/.../tests/ permanently (no graduation)

🚨 CRITICAL: BSP numbers repeat at different levels—always use full paths.

21-foo.capability/21-bar.feature/54-baz.story/  ← One story-54
21-foo.capability/37-qux.feature/54-baz.story/  ← DIFFERENT story-54
37-other.capability/21-bar.feature/54-baz.story/  ← DIFFERENT story-54

ALWAYS use the FULL PATH when referencing work items:

Why this matters:

• BSP numbers encode dependency order within each level, not global uniqueness
• Different capabilities can have identically-numbered features
• Different features can have identically-numbered stories
• Without the full path, agents will access the WRONG work item

When communicating about work items:

1. Always include the full hierarchy path
2. Never use bare numbers like "story-54" without context
3. When in doubt, use the absolute path from spx/

Lower BSP = dependency. Higher-BSP items may rely on lower-BSP items.

Check dependencies before working on an item—lower-numbered items in the same scope may need to be done first.

This applies at every level:

1. List all work items in BSP order (capability → feature → story)
2. Return the FIRST item where status ≠ DONE
3. That item blocks everything after it

Example:

48-test-harness.feature/ [OPEN]        ← Was added after 87 but blocks it
87-e2e-workflow.feature/ [IN_PROGRESS] ← Was already started, then dependency discovered

Next work item: 48-test-harness.feature/ → its first OPEN story.

Use position 21 (leaves room for ~10 items before/after):

# First feature in a new capability
21-foo.capability/
└── 21-first-feature.feature/

Use midpoint: new = floor((left + right) / 2)

# Insert between 21 and 54
new = floor((21 + 54) / 2) = 37

21-first.feature/
37-inserted.feature/    ← NEW
54-second.feature/

Use midpoint to upper bound: new = floor((last + 99) / 2)

# Append after 54
new = floor((54 + 99) / 2) = 76

21-first.feature/
54-second.feature/
76-appended.feature/    ← NEW

1. Directory: NN-{slug}.{type}/ (e.g., 21-auth.capability/)
2. Spec file: {slug}.{type}.md (e.g., auth.capability.md)
3. Tests directory: tests/ (create when starting work)

Optional:

• Decision Records: NN-{slug}.adr.md or NN-{slug}.pdr.md (interleaved with work items)

${SKILL_DIR}/templates/decisions/architectural-decision.adr.md

Read the template and adapt:

# Read ADR template
Read: ${SKILL_DIR}/templates/decisions/architectural-decision.adr.md

# Adapt for your decision
- State the architectural concern this decision governs (atemporal voice — no history)
- Document the chosen option and rationale
- Specify consequences and trade-offs
- Define compliance criteria

ADRs are interleaved with work items at any level:

• Product: spx/NN-{slug}.adr.md
• Capability: spx/NN-{slug}.capability/NN-{slug}.adr.md
• Feature: spx/.../NN-{slug}.feature/NN-{slug}.adr.md

Stories inherit decisions from parent feature/capability.

Format: NN-{slug}.adr.md

• NN: BSP number in range [10, 99]
• slug: Kebab-case description (e.g., use-postgresql-for-persistence)

Lower BSP = dependency. Higher-BSP ADRs may rely on lower-BSP decisions.

ADRs follow the same BSP numbering as work items:

Use position 21 (leaves room for ~10 items before/after):

21-first-decision.adr.md

Use midpoint: new = floor((left + right) / 2)

# Insert between 21 and 54
new = floor((21 + 54) / 2) = 37

21-type-safety.adr.md
37-inserted-decision.adr.md    ← NEW
54-cli-framework.adr.md

Use midpoint to upper bound: new = floor((last + 99) / 2)

# Append after 54
new = floor((54 + 99) / 2) = 76

21-type-safety.adr.md
54-cli-framework.adr.md
76-appended-decision.adr.md    ← NEW

Scope boundaries: ADRs are scoped to product/capability/feature.

Within scope: Lower BSP = dependency that higher-BSP items may rely on.

Cross-scope dependencies: Must be documented explicitly in the ADR content.

Problem: Sequential numbering (01, 02, 03) cannot accommodate discovered dependencies.

Example: You have decisions adr-01, adr-02, adr-03. You discover adr-02 needs a prior decision about type safety. With sequential numbering, you must renumber all subsequent ADRs.

Solution: BSP numbering allows insertion at any point using midpoint calculation.

Problem: If ADR file is renumbered (e.g., adr-23 → adr-37), content with embedded numbers becomes stale.

Examples:

• Header # ADR 23: Foo - wrong after renumbering
• Reference "See ADR-23" - wrong after renumbering

Solution:

• Header: # ADR: Foo (document type prefix, no number)
• References: [Foo](37-foo.adr.md) (markdown link with path)
• Filenames can be renamed, slugs stay stable, markdown links update automatically

Problem: Plain text references like "ADR-23" or even `23-foo.adr.md` break when files are renumbered.

Solution: Markdown links [Decision Title](NN-slug.adr.md):

• Modern editors update links automatically on file rename
• Links are clickable for navigation
• Slug provides stability even if number changes
• Can search by slug if link breaks

${SKILL_DIR}/templates/decisions/product-decision.pdr.md

Read the template and adapt:

# Read PDR template
Read: ${SKILL_DIR}/templates/decisions/product-decision.pdr.md

# Adapt for your decision
- State the product behavior this decision governs (atemporal voice — no history)
- Describe the chosen product behavior
- List product invariants users can rely on
- Specify compliance criteria for product behavior

PDRs are interleaved with work items at any level:

• Product: spx/NN-{slug}.pdr.md
• Capability: spx/NN-{slug}.capability/NN-{slug}.pdr.md
• Feature: spx/.../NN-{slug}.feature/NN-{slug}.pdr.md

Stories inherit product decisions from parent feature/capability.

Format: NN-{slug}.pdr.md

• NN: BSP number in range [10, 99]
• slug: Kebab-case description (e.g., simulation-lifecycle-phases)

When to use which:

• Technical implementation choice → ADR
• User-facing behavior choice → PDR
• If unsure, ask: "Is this about HOW we build it (ADR) or WHAT users experience (PDR)?"

See [Type Safety](21-type-safety.adr.md) for validation approach.

<!-- Feature ADR referencing capability ADR -->

This decision builds on [Config Loading](../21-config-loading.adr.md).

<!-- Story referencing feature ADR -->

Implementation follows [CLI Structure](../21-cli-structure.adr.md).

<!-- From story to capability ADR -->

Architectural constraints: [Commander Pattern](../../21-commander-pattern.adr.md)

<!-- From feature to product ADR -->

Type system: [Type Safety](../../../21-type-safety.adr.md)

❌ Plain text reference: "See ADR-21" or "See PDR-21" ❌ Code-only reference: `21-type-safety.adr.md` or `21-lifecycle.pdr.md` ❌ Number-only reference: "ADR 21 specifies..." or "PDR 21 specifies..."

• Clickable: Navigate directly in editors/viewers
• Stable: Slug provides stability even if number changes
• Updatable: Modern editors can update links on file rename
• Descriptive: Title provides context without opening file

Location: ${SKILL_DIR}/templates/product/product.prd.md

Purpose: Product requirements - user value, customer journey, measurable outcomes

Usage:

# Read PRD template
Read: ${SKILL_DIR}/templates/product/product.prd.md

# Adapt for product change
- Define user value proposition
- Document measurable outcomes with targets
- Specify acceptance criteria
- Avoid implementation details

Placement:

• Product-wide: spx/{product-name}.prd.md
• Capability catalyst: spx/NN-{slug}.capability/{topic}.prd.md

• Immutable: Code adapts to requirements, not vice versa
• Optional catalyst: PRD may exist as enrichment; spec file is always required
• No TRDs: Technical details belong in feature.md, not separate documents

${SKILL_DIR}/templates/outcomes/capability-name.capability.md
${SKILL_DIR}/templates/outcomes/feature-name.feature.md
${SKILL_DIR}/templates/outcomes/story-name.story.md

# For capability
Read: ${SKILL_DIR}/templates/outcomes/capability-name.capability.md
Adapt: Replace {slug} with kebab-case name
       Fill functional requirements
       Add user value context

# For feature
Read: ${SKILL_DIR}/templates/outcomes/feature-name.feature.md
Adapt: Replace {slug} with kebab-case name
       Specify integration scope
       Define component interactions

# For story
Read: ${SKILL_DIR}/templates/outcomes/story-name.story.md
Adapt: Replace {slug} with kebab-case name
       Detail atomic implementation
       List specific functions/classes

Work items follow this pattern:

spx/{BSP}-{slug}.{type}/{slug}.{type}.md

Examples:

• spx/21-core-cli.capability/core-cli.capability.md
• spx/21-core-cli.capability/10-init.feature/init.feature.md
• spx/21-core-cli.capability/15-init.feature/87-parse-flags.story/parse-flags.story.md

Tests stay co-located with their specs permanently. Test level is indicated by filename suffix (.unit.test.*, .integration.test.*, .e2e.test.*) but is not tied to hierarchy level — any test level can appear at any node. See /testing for how to determine the right level.

• Templates exist and are readable in templates/ subdirectories
• Other skills can successfully read templates from appropriate sections
• Progressive disclosure guides readers to relevant sections
• ADR, PDR, requirement, and work item patterns are clearly documented
• Test co-location paths are correctly specified for each level
• BSP numbering uses consistent two-digit format (10-99)