Borrowing it
Nothing to install: this file belongs to Ohh-889/skyroc. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Ohh-889/skyroc/master/.agents/skills/feature-dev/SKILL.mdgit clone --depth 1 https://github.com/Ohh-889/skyrocWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ohh-889/skyroc/feature-dev)<a href="https://agentmods.dev/skills/ohh-889/skyroc/feature-dev"><img src="https://agentmods.dev/badge/skills/ohh-889/skyroc/feature-dev/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ohh-889/skyroc/feature-dev"><img src="https://agentmods.dev/badge/skills/ohh-889/skyroc/feature-dev.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00076 | $0.01097 |
| Opus 5.5 | $0.00030 | $0.00439 |
| Sonnet 5.5 | $0.00015 | $0.00219 |
| Haiku 4.5 | $0.00008 | $0.00110 |
Grade A, and why
feature-dev scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
复杂功能开发
在充分理解现有代码的基础上完成复杂功能。遵循仓库根目录及相关子目录中的 AGENTS.md,将其作为编码规范的唯一事实来源。
工作原则
- 先追踪真实调用链、数据流、模块边界和相邻实现,再确定方案。
- 优先复用现有组件、Hooks、服务、类型和基础设施。
- 默认选择满足需求的最小完整改动,避免无关重构和预设式抽象。
- 只有缺失信息会明显改变业务结果、兼容性、数据安全或实现边界时才询问用户;轻微歧义根据现有代码合理判断并继续。
- 只在方案存在实质性差异时比较多个架构,不为形式完整而制造备选方案。
- 未经用户授权,不扩大任务范围或执行破坏性、不可逆操作。
1. 判断复杂度
先判断任务是否确实需要本工作流。
- 若任务范围清晰、局限于少量文件且不涉及重要架构决策,直接按常规开发流程完成,不强制执行后续阶段。
- 若涉及多个模块或系统、公共接口、数据迁移、复杂状态流或高风险兼容性,继续执行以下流程。
- 仅当任务包含三个以上相互依赖的阶段且计划能帮助跟踪进度时,使用当前环境提供的计划工具。
2. 探索代码库
- 阅读相关
AGENTS.md、包配置、架构说明和相邻实现。 - 追踪入口、调用链、状态变化、数据契约、错误处理和测试方式。
- 搜索可复用的类似功能与扩展点。
- 当存在多个可独立调查的问题且并行探索能明显提高效率时,按需启动探索代理;不要固定代理数量。要求代理返回结论、证据和关键文件列表,随后亲自阅读决定实现所必需的文件。
- 向用户说明会影响实现选择的关键发现;不要输出冗长的文件阅读流水账。
3. 补齐关键信息
根据代码证据识别真正阻塞实现的问题,例如业务边界、错误语义、权限、兼容性、迁移策略和不可逆操作。
- 能从源码、配置、类型、测试或现有行为中确定的信息,直接查明,不反问用户。
- 存在安全且符合项目惯例的默认方案时,说明必要假设并继续。
- 只有不同答案会造成明显不同的业务结果或风险时,集中提出少量具体问题并等待回答。
4. 确定方案
定义目标结构、职责边界、交互方式和兼容策略。
- 通常直接给出一个推荐方案及理由。
- 只有多个方案存在实质性取舍时,简要比较最小改动、长期可维护性和实施风险。
- 若推荐方案仍处于用户已授权的需求范围内,可直接实现。
- 只有方案会扩大范围、改变公共行为、引入数据风险或需要不可逆操作时,才请求用户确认。
5. 实现与验证
- 按选定方案实施最小且完整的修改。
- 严格遵循现有架构、命名、类型和平台规范。
- 保持公共接口和既有行为稳定,除非需求明确要求改变。
- 运行受影响 workspace 的针对性测试、类型检查、lint 或构建;避免默认执行整个大型仓库的全部检查。
- 分析验证失败,区分本次改动、环境问题和已有问题,不把未执行或失败的检查描述为通过。
6. 质量审查
检查功能正确性、边界情况、重复逻辑、抽象边界、项目规范和回归风险。
- 对高风险或跨模块改动,可按需并行委派彼此独立的审查方向;不要固定启动多个审查代理。
- 直接修复任务范围内明确且安全的问题。
- 无关问题只简短记录,不顺手扩大修改范围。
- 需要新增授权或会改变业务行为的问题,向用户说明后再处理。
7. 交付
简洁说明:
- 修改了什么。
- 关键架构或根因是什么。
- 执行了哪些验证以及结果。
- 仍存在的风险或未验证项。
不要生成用户未要求的计划文档、迁移指南或冗长过程复盘。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 79 lines · 76 tokens per session scan A d4deef8233ef
feature-dev is a skill published in the GitHub repository Ohh-889/skyroc (797 stars, last pushed 1mo ago), licensed MIT. It adds 76 tokens to every session and 1,097 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
antd
Use when the user's task involves Ant Design (antd) — writing antd components, debugging antd issues, querying antd APIs/props/tokens/demos, migrating between antd versions, or analyzing antd usage in a project. Triggers on antd-related code, imports from 'antd', or explicit antd questions.
pro-upgrade
An upgrade guide for projects built with Ant Design Pro, a React-based application template for business software. It compares the project with the latest official template and merges framework updates while preserving application code.
react
Use when building or reviewing a client-side React SPA bundled by Vite (React 19.2): components, where state lives, hooks, server data fetching, client routing, re-render and effect bugs. NOT App Router / server components / SSR (that is nextjs); NOT native screens (that is react-native).
v3-upsert-route
A route configuration helper for projects that keep their routes in `src/router/index.ts`. It creates or updates permanent and permission-controlled pages, including titles, icons, and child routes.
v3-use-composables
Documentation for reusable Vue functions in a project, covering device detection, asynchronous dropdown data, loading screens, pagination, routing, themes, page titles, and watermarks.
owl-admin-devtools-generator
A development guide for Owl Admin, a Laravel-based administration system, covering its code generator, API templates, relationships, and visual pages.