codeguard-metal

Metal Codeguard Gate

Goal: turn "looks fine" into repeatable tool evidence, and honestly distinguish pass, fail, unverifiable, and planned.

Start in 30 seconds

Currently planned only means Codeguard can identify .metal, not that an executable gate exists.

1. Clearly report "identified, but currently cannot be verified by Codeguard".
2. Retain the project's existing IDE, compiler, or platform diagnostic results; do not fabricate generic commands.
3. If checking is required, use the target platform's official tools and mark the results as external evidence.
4. Only after detection, lint, failure samples, and regression tests are completed can the registry be upgraded to stable/beta.

Capability boundaries

✅ Good at

1. Identifying Metal files and related user intent.
2. Clearly stating the current Codeguard support status and gaps.
3. Defining verifiable acceptance criteria for future integration.

⚠️ Requires conditions

1. Real target tool or platform diagnostic output.
2. At least one failure sample and one pass sample.
3. A repeatable non-interactive command or a clear host boundary.

❌ Not applicable

1. Do not execute nonexistent generic linter commands.
2. Do not write extension identification as a passed quality gate.
3. Do not upgrade support status without runtime evidence.

When to use

• The user requests checking or fixing Metal code standards.
• AI modified .metal, requiring pre-commit verification.
• CI shows failures related to Metal 工具链 .
• Need to determine whether an issue can be fixed automatically or must be handled manually.

When not to use

• The main task is learning language syntax, framework design, or business modeling; use the corresponding language professional skill instead.
• The main task is dependency vulnerabilities, security auditing, or license governance; use a dedicated security skill.
• When the user only asks to explain one error and there is no project context, first do a localized explanation and do not claim the repository is all green.

Data and security

This skill does not collect, upload, send, or store user code and credentials. By default, it only reads local repository configuration and runs local tools; it does not install dependencies over the network and does not record tokens, passwords, or private source code. Before any automatic fix, check the diff first; destructive or large-scale changes must be explicitly confirmed by the user.

Execution contract

Item	Current fact
Status	planned(since Codeguard V0.5)
File extension	.metal
Project marker	No dedicated marker file
Integration configuration	Based on the repository's existing configuration
Tool readiness	No standalone CLI linter (relies on platform IDE diagnostics)
Detection command	使用命令查找与项目配置检查
lint	未接入
format/fix	未接入
Project-level gate	未接入

Standard Workflow

Step 1: Determine the real scope

Confirm the repository root, module roots, generated directories, vendor directories, and user-specified files. For multi-module projects, first list the modules that will be scanned; do not rely on guessing from the current directory.

Step 2: Read the project contracts

Read 以仓库现有配置为准 and the actual commands in CI. Repository configuration takes precedence over this skill's examples; if the two conflict, report the differences first.

Step 3: Detect tools and versions

Run 对应工具 --version. If a tool is missing, versions are incompatible, or configuration cannot be loaded, record it as "unable to verify", not as passing.

Step 4: Perform read-only checks

Run 目标平台官方诊断, and record the command, working directory, exit code, tool version, and first actionable error.

Step 5: Failure classification

Type	Typical signal	Handling
Formatting	Indentation, whitespace, import order	Formatter allowed, then check the diff
Static rules	unused, complexity, naming, dangerous APIs	Minimal semantic fixes; no bulk guessing
Configuration	parser/config/schema not found	Fix the integration or report the prerequisites
Environment	command not found, version conflict	Report installation/version requirements; do not install silently
Tool internal error	crash, timeout, parser exception	Preserve the original evidence, narrow the reproduction scope

Step 6: Controlled fixes

Prefer running 人工最小修复. Only handle issues that are certain and reversible; dependency upgrades, rule exemptions, generated files, and business semantic changes must be explained separately.

Step 7: Verification loop

Re-run the same lint command, then run the project's existing tests and build. Only write "pass" when the command, scope, and exit code are all explicit; otherwise write "not run" or "unable to verify".

Rules

1. Re-verify with the same command: After fixing, you must re-run the original command that triggered the failure.
2. Layered evidence: Report lint, format, compile, test, and security separately.
3. Minimal changes: Do not refactor along the way, and do not expand stylistic fixes into business rewrites.
4. Configuration first: Do not override the project's existing ignore, dialect, target, or version constraints.
5. No fabrication: Tools not run, configurations not seen, and platforms not verified must be clearly marked.

Output template

Metal Codeguard 结果
- 范围：<仓库/模块/文件>
- 工具：<名称与版本>
- 命令：<实际命令>
- 状态：PASS / FAIL / UNVERIFIED / PLANNED
- 发现：<按格式/规则/配置/环境分类>
- 修改：<文件与原因；无修改写 none>
- 复验：<命令、退出码>
- 未验证：<测试/构建/平台差异>

Gotchas

1. A missing tool is not a pass — you must output UNVERIFIED and a clear preparation method.
2. A formatter is not a linter — all-green formatting cannot cover static rules or compilation errors.
3. The root directory determines the result — running in the wrong module may miss issues or load the wrong configuration.
4. Placeholders are not literal arguments — {file} must be replaced with real, authorized paths.
5. Automatic fixes may expand the diff — after fixing, review the diff first, then run the regression gates.
6. Generated and vendor directories need an explicit policy — Do not arbitrarily scan or modify third-party content across the entire repository.

Domain pitfalls and fixes

1. Domain pitfalls — The current entry is marked as planned; it can only be upgraded after evidence from the Xcode/metal compilation chain is available.

When information is insufficient

Do not only say "please provide more information." First provide a safe read-only inspection plan, and list the specific items that still need confirmation:

1. Repository/module root directory;
2. Target tools and versions;
3. Project configuration files;
4. Authoritative commands in CI;
5. Whether automatic fixes are allowed.

FAQ

Q1: If the tool is not installed, can it be determined that the code has no issues? No. The status must be UNVERIFIED。

Q2: Can ignore or suppression be added automatically? No. An exemption can only be granted when the user explicitly accepts it and the reason is recorded.

Q3: Why does it still fail after formatting? The formatter only covers formatting; static rules, configuration, compilation, and tests still need to be handled separately.

Q4: Should generated code be scanned? By default, follow the repository configuration; if there are no rules, first exclude generated/vendor code, then explain to the user.

Q5: How can too many changes at once be avoided? First fix in batches by failure category and file, and rerun the original command and review the diff after each batch.

Q6: If it passes locally, can it be said that CI will pass? No. It is also necessary to align with CI's versions, environment variables, operating system, and command scope.

Load resources on demand

• When encountering exit codes, failure classifications, or exemption decisions, read references/rules/gate-rules.md。
• When it is necessary to check this language's registry, toolchain, and project configuration, read references/tooling/toolchain.md。
• When copyable task examples are needed, read by scenario examples/, and it is not necessary to load all of them at once.