在离线 Windows 环境中实现项目功能改动
Scanned 9/2/2026
Install to Claude Code
npx -y skills add wangzhe04/ruyi-workbench-oss --skill feature-development --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Feature Development?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/wangzhe04-feature-development)More formats (shields.io, HTML) on the badges page.
---
name: 功能开发
description: 在离线 Windows 环境中实现项目功能改动
---
# Feature Development
Use this skill when implementing a project change in an offline Windows Claude Code environment.
Workflow:
1. Use `project_snapshot`, `dependency_inventory`, and `docs_search` to understand the project before editing.
2. Check `git_status` so user work is visible and unrelated changes are not overwritten.
3. Make a short implementation plan tied to files and tests.
4. Edit narrowly, following local style and existing helpers.
5. Run the most relevant local tests, build, or smoke check with `powershell_run` or `script_run`.
6. Finish with changed files, verification results, and any remaining manual step.
Offline rules:
- Do not require npm, pip, cargo, or NuGet downloads at runtime unless the package cache is already bundled.
- Prefer vendored docs and local README files via `docs_search`.
- If a missing dependency blocks the work, report the exact package/runtime that must be added to the offline bundle.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...