PROACTIVELY USE when planning or executing projects. Auto-discovers codebase patterns, creates multi-phase plans, and orchestrates parallel agent execution. Primary entry point for all planning and execution workflows.
/plugin marketplace add Git-Fg/thecattoolkit/plugin install sys-builder@cattoolkitThis skill is limited to using the following tools:
references/context-management.mdreferences/plan-format.mdreferences/templates/brief.mdreferences/templates/discovery.mdreferences/templates/handoff.mdreferences/templates/issue-entry.mdreferences/templates/issues.mdreferences/templates/phase-plan.mdreferences/templates/roadmap.mdreferences/templates/summary.mdComprehensive planning and execution system with automatic codebase discovery and optimized parallel agent orchestration. This is the primary entry point for all project planning and execution workflows.
MANDATORY: All planning artifacts MUST be created in:
.cattoolkit/planning/{project-slug}/
NEVER create PLAN.md, ROADMAP.md, or BRIEF.md at the project root.
| Mode | Entry Point | Description |
|---|---|---|
| Plan | User requests planning | "Create a plan for X", "I need a roadmap for Y" |
| Execute | User requests execution | "Execute this plan", "Run phase 02" |
| Lite | Simple features | "Add X feature" (2-3 tasks, single phase plan file) |
1. Context Assessment Protocol
Before executing ANY plan, run: find . -maxdepth 2 -not -path '*/.*' -type f | wc -l
2. Execution Mode Selection
| File Count | Complexity | Mode | Cost | Action |
|---|---|---|---|---|
| < 10 | Low | INLINE | 1 | Execute directly in current session. |
| 10 - 50 | Medium | INLINE | 1 | Use Read + Grep to navigate. Do NOT spawn agent. |
| > 50 | High | AGENT | 2xN | Spawn @director (or specific worker) to handle scale. |
3. Anti-Pattern Warning
Spawning a "Planner" or "Director" agent for a project with < 50 files is a Quota Violation. You are smart enough to handle small contexts directly.
4. Template Access
Load references/templates/ on-demand based on Plan Type.
@director (Forked when >50 files OR >5 subdirectories)| Document | Template | Purpose |
|---|---|---|
| Project Brief | references/templates/brief.md | Project definition |
| Roadmap | references/templates/roadmap.md | Multi-phase overview |
| Phase Plan | references/templates/phase-plan.md | Executable phase plans |
| Handoff | references/templates/handoff.md | Session pause/resume |
| Summary | references/templates/summary.md | Phase completion |
| Discovery | references/templates/discovery.md | Auto-discovery findings |
| Issues | references/templates/issues.md | Deferred enhancements |
Create the planning directory:
mkdir -p .cattoolkit/planning/{slug}/phases
Output Structure:
.cattoolkit/planning/{slug}/
├── BRIEF.md # Project definition
├── DISCOVERY.md # Auto-discovery findings
├── ROADMAP.md # Phases with parallelism markers
└── phases/
├── 01-{name}/
│ └── 01-01-PLAN.md (phase plan file)
└── 02-{name}/
└── 02-01-PLAN.md (phase plan file)
Context Assessment Protocol:
Check Project Size: Run find . -maxdepth 2 -not -path '*/.*' -type f | wc -l
Decision Point:
Parallel Agent Protocol (Large Projects Only):
Launch 2-3 explore agents in parallel via Task tool. Each agent focuses on different aspects:
Synthesize findings into DISCOVERY.md using references/templates/discovery.md.
Lite Plan: Simple feature (2-3 tasks). Create phase plan file directly in .cattoolkit/planning/{project-slug}/phases/XX-name/.
Standard Plan: Complex project. Create full hierarchy:
BRIEF.md (Project goals)ROADMAP.md (Phases and milestones).cattoolkit/planning/{project-slug}/phases/XX-name/Use templates from references/templates/:
BRIEF.md (using brief.md template):
ROADMAP.md (using roadmap.md template):
Phase plan file (using phase-plan.md template):
Create ROADMAP.md with explicit parallelism markers:
| Phase | Name | Status | Parallel Agents | Dependencies |
|-------|------|--------|-----------------|--------------|
| 01 | Discovery & Setup | pending | 2 explore | none |
| 02 | Core Implementation | pending | 3 worker | 01 |
| 03 | Integration | pending | 2 worker | 02 |
State-in-Files Enforcement:
Before proceeding, check for interruption state:
.cattoolkit/planning/[phase-name]/HANDOFF.mdClassify each task:
Sequential Tasks:
Parallel Tasks:
Background Tasks:
run_in_background: trueParallel Execution Pattern:
For parallel tasks 2, 3, 4:
Agent Prompt Template:
You are executing Task {N} of Phase {XX}.
## Context
{Include BRIEF.md summary}
{Include relevant DISCOVERY.md sections}
## Your Task
**Scope**: {files/directories}
**Action**: {what to do}
**Verify**: {how to verify}
**Done**: {acceptance criteria}
## Execution Rules
- Execute in UNINTERRUPTED FLOW mode
- Self-verify before completing
- Log any blockers to HANDOFF.md
- Do NOT ask user questions during execution
## Output
When complete, report:
- Files modified
- Verification results
- Any issues encountered
Verification Protocol:
After each task:
After each phase:
On blocker:
Sequential Tasks:
Parallel Tasks:
Background Tasks:
run_in_background: true| Phase Type | Recommended Agents |
|---|---|
| Discovery | 2-3 explore |
| Core Implementation | 3 worker |
| Integration | 2 worker |
| Testing | 2 worker |
| Review | 2-3 reviewer |
This skill should be used when the user asks to "create a hookify rule", "write a hook rule", "configure hookify", "add a hookify rule", or needs guidance on hookify rule syntax and patterns.
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.