Load when executing spec/design/plan/implement commands for a specific feature. Provides templates, agent dispatch patterns, and workflow orchestration for per-feature work.
Orchestrates the complete per-feature workflow from specification through implementation. It triggers on commands like `/katachi:spec-feature <ID>` to manage document creation, agent dispatch for reviews, and status tracking in FEATURES.md.
/plugin marketplace add asermax/claude-plugins/plugin install katachi@asermax-pluginsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
references/design-template.mdreferences/feature-design.mdreferences/feature-spec.mdreferences/implementation-plan.mdreferences/plan-template.mdreferences/spec-template.mdOrchestrates the per-feature workflow from spec through implementation.
Load this skill when executing:
/katachi:spec-feature <ID>/katachi:design-feature <ID>/katachi:plan-feature <ID>/katachi:implement-feature <ID>/katachi:retrofit-design <ID> (retrofit mode)Guidance documents (how to write each document type):
references/spec-template.md - How to write feature specificationsreferences/design-template.md - How to write design rationalereferences/plan-template.md - How to write implementation plansDocument templates (actual templates to follow):
references/feature-spec.md - Feature specification templatereferences/feature-design.md - Design document templatereferences/implementation-plan.md - Implementation plan templateBefore starting any per-feature command:
# Check feature exists in FEATURES.md
feature = get_feature(FEATURE_ID)
if not feature:
error("Feature not found in FEATURES.md")
# Check dependencies are complete (for design/plan/implement)
if command in ["design", "plan", "implement"]:
deps = get_dependencies(FEATURE_ID)
incomplete = [d for d in deps if not is_complete(d, required_phase)]
if incomplete:
warn(f"Dependencies not complete: {incomplete}")
Update status when starting:
# spec-feature
python scripts/features.py status set FEATURE-ID "⧗ Spec"
# design-feature
python scripts/features.py status set FEATURE-ID "⧗ Design"
# plan-feature
python scripts/features.py status set FEATURE-ID "⧗ Plan"
# implement-feature
python scripts/features.py status set FEATURE-ID "⧗ Implementation"
For spec/design/plan commands:
Research Phase (Silent)
Draft Proposal
Present for Review
Iterate
Validate
Finalize
Each command dispatches its reviewer agent:
| Command | Agent | Input |
|---|---|---|
| spec-feature | katachi:spec-reviewer | Feature description, completed spec |
| design-feature | katachi:design-reviewer | Spec, design, ADR/DES summaries |
| plan-feature | katachi:plan-reviewer | Spec, design, plan, ADR/DES summaries |
| implement-feature | katachi:code-reviewer | Spec, design, plan, code, ADR/DES |
| retrofit-design | katachi:codebase-analyzer, katachi:design-reviewer | Spec, implementation code, ADR/DES indexes |
Dispatch pattern:
Task(
subagent_type="katachi:spec-reviewer",
prompt=f"""
Review this feature specification:
## Feature Description (from FEATURES.md)
{feature_description}
## Completed Spec
{spec_content}
Provide structured critique following your review criteria.
"""
)
Update status when completing:
# After successful completion
python scripts/features.py status set FEATURE-ID "✓ Spec"
python scripts/features.py status set FEATURE-ID "✓ Design"
python scripts/features.py status set FEATURE-ID "✓ Plan"
python scripts/features.py status set FEATURE-ID "✓ Implementation"
The implementation workflow is more autonomous:
Read all documentation silently
Implement all steps autonomously
Validate with code-reviewer
Present for user review
Iterate based on feedback
Surface patterns for DES
During implementation, watch for:
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.