codeguard-haskell

Haskell Codeguard Gate

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

Start in 30 seconds

1. At the repository root, confirm stack.yaml`, `cabal.project or the target file exists.
2. Detect tools:hlint --version。
3. First do a read-only check:hlint .。
4. Classify results as: code violations, missing configuration, missing tools, tool internal errors.
5. Only perform on safely fixable items:fourmolu -i .。
6. Re-run lint, and run the project's existing test/build gates.

You can directly ask:

• "Check this Haskell project, don't modify anything yet, and give me a classification of lint failures."
• "Only fix Haskell formatting issues; list semantic issues first."
• "Verify Haskell lint, test, and build evidence before committing."

Capability Boundaries

✅ Good at handling

1. Based on .hs, .lhs and project markers to identify the Haskell workspace.
2. Run hlint . and preserve exit codes, scope, and tool versions.
3. Distinguish formatting, static rules, configuration, dependency, and environment failures.
4. Use fourmolu -i . to handle reversible issues and re-scan.
5. Report local results and CI/build/test evidence in layers.

⚠️ Requires conditions

1. The target tool must be installed and its version compatible with project constraints.
2. Must be executed from the correct repository or module root directory.
3. If the project has custom configuration, the committed configuration prevails.
4. Whether generated code, vendor, and cache directories are scanned must follow repository rules.

❌ Not applicable

1. Do not treat formatter passing as compilation, test, or security audit passing.
2. Do not silently install tools, upgrade dependencies, or modify team rules.
3. Do not use suppress/ignore/raising thresholds to hide real problems.
4. Do not claim "check passed" when tools or configuration are missing.

When to use

• The user requests checking or fixing Haskell code style.
• AI modified .hs, .lhs, and pre-commit verification is needed.
• CI has failures related to hlint . .
• Need to determine whether a problem can be fixed automatically, or must be handled manually.

When not to use

• When the main task is learning language syntax, framework design, or business modeling, use the corresponding language professional skill instead.
• When 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 give a local 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	stable(since Codeguard V0.5)
File Extensions	.hs, .lhs
Project Markers	stack.yaml, cabal.project
Integration Configuration	Subject to the repository's existing configuration
Tool Preparation	brew install hlint fourmolu
Detection Commands	使用命令查找与项目配置检查
lint	hlint .
format/fix	fourmolu -i .
Project-level gate	hlint .

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 guess based on the current directory.

Step 2: Read the project contract

Read the actual commands in the repository's existing configuration and in CI. Repository configuration takes precedence over the examples in this skill; if the two conflict, report the differences first.

Step 3: Detect tools and versions

Run 对应工具 --version. Missing tools, incompatible versions, or configurations that fail to load are all recorded as "unable to verify", not as passing.

Step 4: Run read-only checks

Run hlint ., recording the command, working directory, exit code, tool version, and the 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 errors	crash, timeout, parser exceptions	Preserve the original evidence and narrow the reproduction scope

Step 6: Controlled fixes

Prefer to run fourmolu -i .. 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 style 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 explicitly marked.

Output Template

Haskell 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 an explicit preparation method.
2. A formatter is not a linter — all-green formatting cannot cover static rule or compilation errors.
3. Root directory determines the result — Running in the wrong module may miss findings or load the wrong configuration.
4. Placeholders are not literal arguments — {file} must be replaced with real, authorized paths.
5. Automatic fixes may enlarge the diff — After fixing, review the diff first, then run the regression gates.
6. Generated and vendor directories require an explicit policy — Do not scan or modify third-party content across the whole repository at will.

Domain pitfalls and fixes

1. Domain pitfalls — HLint suggestions are not always semantically equivalent; when strictness or performance is involved, do not apply them blindly in bulk.

When information is insufficient

Do not just say "please provide more information." First give 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. The authoritative commands in CI;
5. Whether automatic fixes are allowed.

FAQ

Q1: If the tool is not installed, can we conclude the code is fine? No. The status must be UNVERIFIED。

Q2: Can we automatically add ignore or suppression? No. Exemptions are allowed only when the user explicitly accepts them 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; when there are no rules, exclude generated/vendor first, then explain to the user.

Q5: How can we avoid changing too much at once? First fix in batches by failure category and file, and rerun the original command and review the diff for each batch.

Q6: If it passes locally, can we say CI will pass? No. You also need to align the CI 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 you need to check this language's registry, toolchain, and project configuration, read references/tooling/toolchain.md。
• When you need copyable task examples, read by scenario examples/, and there is no need to load all of them at once.