cycle
/cycle
Run the full Plan → Build → Verify loop for a SpawnForge feature. Orchestrates planner → builder → validator in sequence. Use when implementing a complete feature end-to-end from spec to passing tests.
- Author
- Tristan578 ↗
- Repository
- project-forge ↗
- Licence
- Apache-2.0
- Last repository change
- 17.10.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 17.10.2026, without references to retired models or APIs.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
What it does and when to use it
Run the full Plan → Build → Verify loop for a SpawnForge feature. Orchestrates planner → builder → validator in sequence. Use when implementing a complete feature end-to-end from spec to passing tests.
Original author description
Run the full Plan → Build → Verify loop for a SpawnForge feature. Orchestrates planner → builder → validator in sequence. Use when implementing a complete feature end-to-end from spec to passing tests.
How it works
- Overview
- 1. Plan
- 2. Build
- 3. Verify
- 4. Update Context
- Version Reference
SKILL.md preview
--- name: cycle description: Run the full Plan → Build → Verify loop for a SpawnForge feature. Orchestrates planner → builder → validator in sequence. Use when implementing a complete feature end-to-end from spec to passing tests. --- # The Development Loop ## Overview This is the full development cycle for SpawnForge. Every feature goes through Plan → Build → Verify → Update Context. Each phase uses domain-specific skills to ensure consistent, high-quality output aligned with the product vision. ## 1. Plan Invoke the `planner` skill to draft a spec for: $ARGUMENTS - The planner uses the `design` skill patterns for architectural decisions - The planner verifies claims against actual code (not just design docs) - **STOP and await user approval.** ## 2. Build Once approved, invoke the `builder` skill to implement it. The builder will: - Use `/rust-engine` patterns for Bevy 0.19 / Rapier 0.35 / WASM code - Use `/frontend` patterns for React 19 / Next.js 16 / Zustand 5 / Tailwind 4 code - Use `/mcp-commands` patterns for AI parity (commands, handlers, manifest) - Use `/testing` patterns for 100% coverage target - Use `/docs` patterns for documentation updates ## 3. Verify Run the architecture validator and test suite: ```bash # Architecture boundaries python3 .claude/skills/arch-validator/check_arch.py # Lint (zero warnings enforced) cd web && npx eslint --max-warnings 0 . # TypeScript cd web && npx tsc --noEmit # Unit tests cd web && npx vitest run # MCP tests cd mcp-server && npx tsc --noEmit && npx vitest run ``` If any check fails, fix immediately before proceeding. ## 4. Update Context After verification passes, update project context files: - **New pitfalls/API quirks?** → Add to `.claude/rules/*.md` - **Milestone completed?** → Move its tickets on the taskboard; update the feature claims in `README.md` if a user-visible capability shipped. There is deliberately no phase list in `.claude/CLAUDE.md` — a hand-maintained status summary rots and the taskboard is the live source - **New ECS components or libraries?** → Update `rules/file-map.md`, `rules/bevy-api.md`, `rules/library-apis.md` - **New EntitySnapshot fields?** → Update `rules/entity-snapshot.md` - **Temporary learnings?** → Log in `MEMORY.md`, promote to rules after confirmation ## Version Reference All implementations must use these exact versions: - Bevy 0.19, bevy_rapier 0.35, bevy_hanabi 0.19, bevy_panorbit_camera 0.35 - Next.js 16, React 19, Zustand 5, TypeScript 6, Tailwind 4, Vitest 5 - Rust stable, wasm-bindgen 0.2.127, csgrs 0.20
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · Tristan578/project-forge | 100 | Apache-2.0 |
| softagram/goodreason-agents-setup | 90 | MIT |
| pedrobertao/4D-framework-enhanced | 90 | MIT |
| dudwns0921/claude-gamedev-kit | 87 | MIT |
Identical content appears in 1 other GitHub folders. We show one copy with the highest score.
Database snapshot: 13.11.2026 · package version a0f76e93a6ca (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 100planner
Create a detailed spec in specs/ for a SpawnForge feature. Use when starting any new feature, asked to plan a complex change, or when "spec", "design", or "architect" appears in a request. Produces implementation-ready specs with acceptance criteria. - 100feature-cycle
End-to-end orchestrator chaining spec → plan → implement → verify → review → ship from a local spec — the whole CLAUDE.md lifecycle as one command. Use to drive a complete feature from a ready spec. To validate what to build first use /office-hours. - 100implement-feature
End-to-end feature implementation workflow. Use when the user says "implement", "build feature", "work on task", "start implementation", or any feature development work. Covers plan review, test writing, implementation, verification, and delivery. - 100ship
Run the full spec-to-merge pipeline as one command: spec, plan, small-batch build, code review, and a PR with auto-merge — pausing at the existing human gates. Idempotent per issue — a re-invocation for work already shipped or in-flight resumes/monitors instead of re-running the… - 100test-plan
Plan and check testing for a feature: map each acceptance criterion to a test at the right level, find gaps, write the missing tests, and run end-to-end checks of the main flows. Use when a feature needs a test strategy or before review. - 100pr-code-review
Review a PR for SpawnForge bugs, security issues, and regressions. Posts inline GitHub comments with severity markers. Use when asked to "review PR", "code review", or "check this PR". Invoke with /pr-code-review <PR number>.