project-aware-coding

project-aware-coding is a skill for Codex from Zhangs-11/zs-skills. It costs 44 tokens per session (2,359 once invoked), scanned A, original, MIT.

A development workflow for changing an existing code repository after tracing how its code, data, tests, and configuration actually work together.

In plain words
What is it for?
It helps implement features, fix bugs, change APIs or schemas, update configuration, and refactor code with a small validated change and a self-review.
Why use it?
It reduces changes that fit the name of a function but break a real caller, consumer, tenant, resource, or existing project rule.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: mentions CLAUDE.md; mentions AGENTS.md; $skill-name invocation.

Good fit It helps implement features, fix bugs, change APIs or schemas, update configuration, and refactor code with a small validated change and a self-review.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/zhangs-11/zs-skills/project-aware-coding
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Zhangs-11/zs-skills --skill project-aware-coding
Clone the repo
git clone --depth 1 https://github.com/Zhangs-11/zs-skills

Made for: Codex.

Wrote 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.

agentmods badge for project-aware-coding

README.md
[![agentmods](https://agentmods.dev/badge/skills/zhangs-11/zs-skills/project-aware-coding.svg)](https://agentmods.dev/skills/zhangs-11/zs-skills/project-aware-coding)
Your own site
<a href="https://agentmods.dev/skills/zhangs-11/zs-skills/project-aware-coding"><img src="https://agentmods.dev/badge/skills/zhangs-11/zs-skills/project-aware-coding.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,359 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00044 $0.02359
Opus 5 $0.00022 $0.01179
Sonnet 5 $0.00009 $0.00472
Haiku 4.5 $0.00004 $0.00236

Measured today against content hash 7f8647f75908, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

project-aware-coding 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 today.

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.

project-aware-coding/SKILL.md · 119 lines

How it starts

The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.

结合项目上下文写代码

目标不是把新代码写得“像项目”,而是在动手前理解项目已经形成的业务约束、真实消费者和历史经验,减少重复踩坑。项目已有实现优先作为参考样本,但仍要判断它是否适用于当前需求、版本和调用链。

边界

  • 只修改用户当前需求必需的代码、配置、测试和文档,不顺手重构或扩展未来能力。
  • 先检查工作区、分支、worktree 和仓库说明,保留其他人或其他会话的未知改动。
  • 删除文件、函数、注释、测试、兼容逻辑或数据前单独说明理由并取得确认。
  • 实现授权不包含 commit、push、部署、创建 PR、外部评论或数据库写入;这些动作需要当前批次的明确授权。
  • 仓库若提供专用开发 Skill 或规范工具,例如 goalfy-coding,按当前任务读取并组合使用,但不得让其自动扩大本次写入范围。

工作流

1. 固定目标与现场

读取适用的 AGENTS.mdCLAUDE.md、仓库级 Skill 和开发文档。检查 git status、当前分支、HEAD、worktree 和已有 diff;涉及远端基线时先 git fetch,仅审查当前未提交改动时不为形式强制刷新远端。

用可观察结果明确本次需求:谁在什么条件下触发,现有行为哪里偏离,完成后谁能看到什么变化。若信息能从代码、配置、测试或关联材料中确认,先自行读取;只有关键选择会实质改变方案时才询问用户。

2. 还原最小完整机制

追踪与改动有关的真实链路:

输入或触发
→ 入口与生产者
→ 状态或数据变化
→ 真实消费者
→ 用户可见结果
→ 失败、重试、恢复或终态

只展开与当前需求有关的上下游。不能根据函数名、字段名或注释猜用途;读取方法体、调用方、被调用方、注册点、配置来源和必要运行时契约。

共享配置还要与构造参数交叉检查:当同一个配置对象或字段被多个构造器、工厂或 wrapper 复用,而调用时又传入 bucket、tenant、region、namespace、资源 ID 等会改变语义的参数,列出全部生产调用方及其真实实参组合,区分“配置属于谁”和“本次 client 操作谁”。至少让默认消费者和一个非默认消费者走通同一验证链;默认实例的测试不能代表其他 bucket、租户或资源。

3. 按风险参考项目已有实现

使用 rg 从同仓库开始寻找最接近的样本,常用线索包括业务字段、接口路径、错误码、注册方法、消费者名称、配置键和测试名称。按以下优先级选择真正有参考价值的代码:

  1. 与本次改动共享真实消费者或协议链路;
  2. 同模块、同业务角色或同一状态生命周期;
  3. 有针对性测试或运行证据;
  4. 当前仍在使用且版本接近;
  5. 历史提交明确记录过踩坑原因。

简单文案、局部条件或单点字段改动,通常参考最邻近的一处实现即可,不做无边界考古。涉及协议、Schema、序列化、数据库约束、跨服务接口、并发、幂等、权限或兼容行为时,扩大到真实消费者、相关测试和必要 git log / git blame,确认项目曾经为何选择当前写法。

参考时回答三个问题:

  • 哪部分约束与当前需求相同,可以沿用;
  • 旧设计解决或规避了什么实际问题;
  • 哪部分场景、版本或消费者不同,不能机械复制。

已有代码是经验,不是绝对规则。若它已过时、有缺陷或不适用于新场景,可以偏离;但要写清差异来自新需求、依赖版本还是消费者能力,并用针对性测试验证。不要因为某写法符合通用标准、看起来更优雅或外部项目流行,就跳过本项目真实链路。

找不到可靠样本时,记录搜索过的范围和关键词,再查真实消费者实现、官方当前契约或执行最小探针。不要为了满足流程强行选择不相关代码,也不要仅因没有先例就阻塞低风险实现。

4. 设计并实施最小改动

涉及新增或调整模块职责、接口与依赖方向时,使用 $codebase-design 检查高内聚、低耦合;普通局部修改不额外加载。

优先复用项目现有入口、抽象、错误处理、命名和测试结构,但只复用当前需求真正需要的部分。每个 diff hunk 都应能对应到需求、已确认根因、必要测试或项目强制工具输出;无法对应的格式化、helper 抽取、依赖升级和未来兼容应移除或先征求授权。

改动 Schema、API 或协议时,区分“标准允许”和“项目真实消费者支持”。例如某个 JSON Schema 关键字在规范中合法,不代表当前模型、SDK、网关或工具注册链路接受;使用与项目相同的消费路径和样例验证,不把单一历史结论错误推广到所有协议或版本。

遵循仓库现有结构完成最小实现,使用当前环境提供的精确编辑工具。不要覆盖未知改动,不修改无关文件,不为让测试变绿而放松断言、跳过测试或隐藏错误。

Read the full file on GitHub · 119 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. today Changed · -12 lines · -113 tokens per session 7f8647f75908
  2. 7d ago First seen · 131 lines · 157 tokens per session scan A e1ee69943b51

Subscribe to this mod's changes

project-aware-coding is a skill published in the GitHub repository Zhangs-11/zs-skills (2 stars, last pushed yesterday), licensed MIT. It adds 44 tokens to every session and 2,359 once invoked, about $0.0002 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories

superpowers-zh

Use when constraining AI coding with Chinese TDD methodology, systematic debugging, code review, and verification workflows. Superpowers-zh: Chinese adaptation of the Superpowers AI-assisted programming skills and methodologies.

znlgis/opengis-skills · 45 tokens

evjs

Develop, migrate, review, document, or debug evjs applications and framework code involving file-based pages, ev.config.ts, server functions, API routes, rendering, plugins, development, builds, and deployment.

afx-team/evjs · 45 tokens

Async Correctness 非同期処理の正しさ検証

A code review check for asynchronous code, meaning work that finishes later through async/await, promises, or chained callbacks. It looks for missing waits, ignored errors, and operations that can run in the wrong order or interfere with each other.

s977043/river-review · 95 tokens

End-to-End Wiring 末端到達・貫通の検証

A review check that follows a new or changed process from its starting point to its final result. It checks whether actions such as validation, notification, measurement, or saving remain connected across all involved parts of the application.

s977043/river-review · 83 tokens

a-unslop-code

Finds what makes source code read as AI-written and points you at the parts that actually ship bugs. Sorts every tell into three buckets and fixes them in that order: bugs (swallowed errors, a made-up API, a left-in "rest of the code" stub), substance (tutorial-shaped boilerplate, over-engineering, code that ignores…

imisic/claude-marketplace · 187 tokens

Failure Modes & Observability in Design

Ensure designs specify failure modes, timeouts, error contracts, and observability for critical flows.

s977043/river-review · 27 tokens