docs: add comprehensive Skill Development Guide#929
Conversation
- Add comprehensive code review standards for all languages - Define when to review (after code changes, before commits) - Include security review triggers and severity levels - Reference relevant agents (code-reviewer, security-reviewer, etc.) - Add review checklist covering security, quality, and performance - Define approval criteria (Approve/Warning/Block) This rule complements the existing code-reviewer agent by providing clear guidelines on when and how to conduct code reviews.
- Add rules/zh/ directory with complete Chinese translations - Translate all 10 common rule files: - coding-style.md - security.md - testing.md - git-workflow.md - performance.md - patterns.md - hooks.md - agents.md - development-workflow.md - code-review.md - Add README.md for the zh directory explaining structure and installation - Maintain consistent formatting with original English versions - Keep technical terms and code examples in English where appropriate
- Add docs/SKILL-DEVELOPMENT-GUIDE.md with detailed guidance on creating skills - Update CONTRIBUTING.md with enhanced skills section linking to the new guide - Covers skill architecture, categories, best practices, testing, and examples The new guide provides: - What skills are and when they activate - Skill file structure and format - Step-by-step skill creation tutorial - Writing effective skill content - Common patterns and anti-patterns - Testing and validation checklist - Complete examples gallery
|
Analysis Failed
Troubleshooting
Retry: |
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughAdds comprehensive documentation: a new Skill Development Guide and updated CONTRIBUTING/SKILL.md guidance (activation, anti-patterns, checklist), a new common code-review rule, and a suite of Chinese-localized rules covering agents, workflows, style, security, testing, hooks, patterns, performance, and git practices. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Greptile SummaryThis PR adds a comprehensive Skill Development Guide ( The documentation is well-structured and clearly written. A few minor consistency issues were found:
Confidence Score: 5/5Documentation-only PR; safe to merge. All findings are P2 style/consistency suggestions. No P0 or P1 issues were found. The three comments are all P2: a teaching inconsistency in an example, a reference to undefined agent names, and a link-text mismatch. None of these block correct functionality or cause data/runtime problems. rules/common/code-review.md — the agent name table references agents not present in agents.md and should be reconciled before contributors try to use the rule. Important Files Changed
Flowchart%%{init: {'theme': 'neutral'}}%%
flowchart TD
A[Contributor wants to create a skill] --> B[Read docs/SKILL-DEVELOPMENT-GUIDE.md]
B --> C{Choose skill category}
C --> D[Language Standards]
C --> E[Framework Patterns]
C --> F[Workflow]
C --> G[Domain Knowledge]
C --> H[Tool Integration]
D & E & F & G & H --> I[Create skills/your-skill-name/SKILL.md]
I --> J[Add YAML frontmatter\nname · description · origin]
J --> K[Write sections\nWhen to Activate · Core Concepts\nCode Examples · Anti-Patterns\nBest Practices · Related Skills]
K --> L[Validate per checklist\nin CONTRIBUTING.md]
L --> M{All checks pass?}
M -- No --> K
M -- Yes --> N[Run code example tests\nnpx tsc / py_compile / go build]
N --> O[Submit PR using\ngit workflow]
O --> P[Code review via\nrules/common/code-review.md]
P --> Q[Merged into ECC]
Reviews (3): Last reviewed commit: "fix(docs): resolve skill guide review is..." | Re-trigger Greptile |
|
|
||
| - Using [API/Library] | ||
| - Looking up [API/Library] syntax | ||
|
|
There was a problem hiding this comment.
Guide exceeds its own stated maximum line count
The guide recommends "800 lines maximum" for skills, but SKILL-DEVELOPMENT-GUIDE.md itself is 905 lines. While this is a documentation guide rather than a skill, the inconsistency may confuse contributors who use this file as a reference. Consider either increasing the stated maximum to reflect a more realistic upper bound (e.g., 1000 lines for especially comprehensive guides) or trimming the guide to stay within the 800-line limit it recommends.
Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!
There was a problem hiding this comment.
9 issues found across 14 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="rules/zh/coding-style.md">
<violation number="1" location="rules/zh/coding-style.md:48">
P3: Markdown file is missing a trailing newline, which can violate markdownlint (MD047) and fail CI lint checks.</violation>
</file>
<file name="rules/zh/testing.md">
<violation number="1" location="rules/zh/testing.md:29">
P2: New Markdown file is missing a trailing newline, which can fail markdownlint checks (MD047) and block CI/review.</violation>
</file>
<file name="rules/zh/development-workflow.md">
<violation number="1" location="rules/zh/development-workflow.md:35">
P2: Workflow documentation omits required pre-review gates (green automated checks and resolved merge conflicts), leading to non-reviewable PRs.</violation>
</file>
<file name="rules/zh/performance.md">
<violation number="1" location="rules/zh/performance.md:55">
P3: File is missing a trailing newline at EOF, which can fail lint/CI checks enforcing end-of-file newlines.</violation>
</file>
<file name="rules/zh/code-review.md">
<violation number="1" location="rules/zh/code-review.md:48">
P2: HIGH-severity handling is internally contradictory: the severity table says HIGH issues should be fixed before merge, but approval criteria allows merging with only HIGH issues.</violation>
</file>
<file name="CONTRIBUTING.md">
<violation number="1" location="CONTRIBUTING.md:167">
P3: Example Skills uses category "Template" which is not defined in the Skill Categories taxonomy, creating an internal inconsistency that can mislead contributors.</violation>
</file>
<file name="rules/zh/README.md">
<violation number="1" location="rules/zh/README.md:27">
P3: README tree for `common/` vs `zh/` is inconsistent with the actual directories and with the claim that `zh/` is a translation of `common/` (missing `development-workflow.md` and `code-review.md` in the common list, and missing `code-review.md` in the zh list).</violation>
<violation number="2" location="rules/zh/README.md:64">
P2: Manual install steps copy into `~/.claude/rules/...` without creating the parent directory, so `cp -r` fails on a clean environment where `~/.claude/rules` doesn't exist.</violation>
</file>
<file name="rules/common/code-review.md">
<violation number="1" location="rules/common/code-review.md:15">
P2: Code review standards omit a required pre-review step to resolve merge conflicts, causing review-blocking PRs to remain compliant with the written rule.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
| @@ -0,0 +1,116 @@ | |||
| # 代码审查标准 | |||
There was a problem hiding this comment.
P2: HIGH-severity handling is internally contradictory: the severity table says HIGH issues should be fixed before merge, but approval criteria allows merging with only HIGH issues.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At rules/zh/code-review.md, line 48:
<comment>HIGH-severity handling is internally contradictory: the severity table says HIGH issues should be fixed before merge, but approval criteria allows merging with only HIGH issues.</comment>
<file context>
@@ -0,0 +1,116 @@
+| 级别 | 含义 | 行动 |
+|-------|---------|--------|
+| CRITICAL(关键) | 安全漏洞或数据丢失风险 | **阻止** - 合并前必须修复 |
+| HIGH(高) | Bug 或重大质量问题 | **警告** - 合并前应修复 |
+| MEDIUM(中) | 可维护性问题 | **信息** - 考虑修复 |
+| LOW(低) | 风格或次要建议 | **注意** - 可选 |
</file context>
There was a problem hiding this comment.
Actionable comments posted: 4
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
CONTRIBUTING.md (1)
477-482:⚠️ Potential issue | 🟠 MajorScope the file-naming rule so it doesn’t conflict with
skills/**/SKILL.md.Line 479 currently reads as a global rule and can be read as contradicting the required uppercase
SKILL.mdconvention for skills. Please explicitly exclude skills in this section.Suggested doc fix
-## File Naming - -- Use lowercase with hyphens: `python-reviewer.md` -- Be descriptive: `tdd-workflow.md` not `workflow.md` -- Match name to filename +## File Naming + +- For agents/commands/docs files, use lowercase with hyphens: `python-reviewer.md` +- For skills, use folder naming `skills/<lowercase-hyphen-name>/` and keep the file name as `SKILL.md` +- Be descriptive: `tdd-workflow.md` not `workflow.md`Based on learnings: In the
skills/directory, the standard convention is uppercaseSKILL.mdinside a lowercase-hyphen skill folder.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@CONTRIBUTING.md` around lines 477 - 482, Update the "File Naming" section to explicitly exempt skill files: clarify that the lowercase-with-hyphens rule (examples `python-reviewer.md`, `tdd-workflow.md`) applies to general docs but does not apply to files under the skills directory; add a short sentence noting the skills convention is different — skills live in folders under skills/ and use UPPERCASE `SKILL.md` inside a lowercase-hyphen skill folder (e.g., skills/my-skill/SKILL.md) so readers won’t interpret the global rule as overriding the skills convention.
🧹 Nitpick comments (11)
CONTRIBUTING.md (1)
148-153: Clarify the line-count requirement wording.Line 152 (
Under 500 lines (800 max)) is ambiguous as a checklist rule. Consider splitting into target vs hard limit to make review criteria objective.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@CONTRIBUTING.md` around lines 148 - 153, The checklist item "Under 500 lines (800 max)" is ambiguous—split it into two explicit rules so reviewers have a clear target and a hard limit; replace that single line with two checklist entries such as "Prefer target: under 500 lines" (or "Target: under 500 lines") and "Hard limit: 800 lines (do not exceed)", and ensure the wording appears in the checklist near the other items so it reads as both a soft target and a firm maximum.rules/zh/security.md (1)
29-29: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/security.md` at line 29, Add a trailing newline to the end of the file containing the line "5. 审查整个代码库中的类似问题" so the file ends with a single newline character (ensure the final line is terminated); this fixes the missing EOF newline for rules/zh/security.md.rules/zh/coding-style.md (1)
48-48: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/coding-style.md` at line 48, 该文件在结尾缺少换行;在 rules/zh/coding-style.md 的末尾(可定位到最后一行包含文本 "[ ] 没有变更(使用不可变模式)")添加一个终止换行符(newline),确保文件以单个换行符结尾并保存改动以通过 POSIX 文件结尾检查。rules/zh/performance.md (1)
55-55: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/performance.md` at line 55, The file ends without a newline after the final line "4. 每次修复后验证"; open the markdown file containing that line and add a single trailing newline character at the end of the file (ensure the file ends with a blank line/line break) so standard tools and POSIX conventions detect the EOF newline correctly.rules/zh/patterns.md (2)
50-50: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/patterns.md` at line 50, The file patterns.md is missing a trailing newline; open the rules/zh/patterns.md file (identify by the filename patterns.md) and add a single newline character at the end of the file so it ends with '\n', then save/commit the change.
31-31: Add newline at end of file.The file is missing a trailing newline, which is a common convention for text files.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/patterns.md` at line 31, The file ends without a trailing newline; open rules/zh/patterns.md and add a single newline character at the end of the file so the last line "包含分页响应的元数据(total、page、limit)" is terminated by a newline (ensure your editor or git commit preserves the final newline).rules/zh/hooks.md (1)
30-30: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/hooks.md` at line 30, The file "rules/zh/hooks.md" is missing a trailing newline after the final line "误解的需求"; open the file and add a single newline character at the end of the file so the file ends with a newline (ensure the final token "误解的需求" is followed by a newline).rules/zh/development-workflow.md (1)
38-38: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/development-workflow.md` at line 38, The file is missing a trailing newline; open the document containing the line "参见 [git-workflow.md](./git-workflow.md) 了解提交消息格式和 PR 流程" and add a single newline character at the end of the file so the file ends with a blank line (ensure the EOF has a newline).rules/zh/git-workflow.md (1)
24-24: Add newline at end of file.The file is missing a trailing newline at the end.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/git-workflow.md` at line 24, The file's final line ("参见 [development-workflow.md](./development-workflow.md).") is missing a trailing newline; open the markdown file and ensure it ends with a single newline character (add an empty line or press Enter after that last line) so the file terminates with a newline.rules/zh/README.md (1)
99-102: Optional: Consider grammar refinement in Chinese.The static analysis tool flagged line 100: the phrase "可能被语言特定文件覆盖" (may be overridden by language-specific files) uses a modal verb in a passive construction that may sound slightly awkward in Chinese. Consider "会被语言特定文件覆盖" (will be overridden) for more natural phrasing, though the current version is semantically appropriate.
📝 Optional grammar refinement
-`rules/common/` 中可能被语言特定文件覆盖的规则会标记: +`rules/common/` 中会被语言特定文件覆盖的规则会标记:🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/README.md` around lines 99 - 102, 将说明文本中的措辞从“可能被语言特定文件覆盖”替换为更自然的“会被语言特定文件覆盖”以改善中文表达;在文件中查找包含该短语(例如在 rules/zh/README.md 中标记规则说明段落或引文区域)并直接替换该片段,保持其后续说明“:此规则可能被语言特定规则覆盖,对于某些语言,该模式可能不是惯用的。”相应一致(可同时将后半句中的“可能不是惯用的”保留或改为“可能不适用”以风格统一)。rules/zh/agents.md (1)
20-27: Consider adding security-reviewer to the immediate usage list.The section covers 4 key scenarios for proactive agent usage but omits
security-reviewer, which appears in the agents table above. Based on learnings, the security-reviewer should be used "before commits to perform security analysis," making it another candidate for proactive/immediate usage guidance.💡 Suggested addition
4. 架构决策 - 使用 **architect** 代理 +5. 安全敏感代码或提交前 - 使用 **security-reviewer** 代理🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@rules/zh/agents.md` around lines 20 - 27, The "立即使用代理" list omits the security-reviewer; update that section (the "## 立即使用代理" block that currently lists planner, code-reviewer, tdd-guide, and architect) to include security-reviewer as an immediate-use agent (e.g., add a new bullet "5. 提交前进行安全分析 - 使用 **security-reviewer** 代理" or similar), ensuring the agent name matches the agents table elsewhere so readers can find its description.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@CONTRIBUTING.md`:
- Around line 160-168: The category mismatch: the row for
`project-guidelines-example/` uses Category "Template" which isn't defined in
the Skill Categories table; either add a "Template" entry to the Skill
Categories table (matching the format used for other entries) or change the
`project-guidelines-example/` row to an existing category (e.g., "Workflow" or
"Domain Knowledge") so the Skill Categories table and examples table stay
consistent; update whichever of the two you modify
(`project-guidelines-example/` or the Skill Categories table) accordingly.
In `@docs/SKILL-DEVELOPMENT-GUIDE.md`:
- Around line 66-101: The SKILL.md canonical template is missing the required
"## Anti-Patterns" section; update the SKILL.md format block in the
SKILL-DEVELOPMENT-GUIDE so it includes a "## Anti-Patterns" heading (with
guidance bullets/examples) consistent with CONTRIBUTING.md; locate the template
block showing headers like "## When to Activate", "## Core Concepts", and "##
Best Practices" and insert a new "## Anti-Patterns" section (between Core
Concepts and Code Examples or next to Best Practices) describing common pitfalls
and examples to follow the repo's contribution standards.
In `@rules/zh/agents.md`:
- Around line 43-50: The file ends immediately after the "## 多视角分析" section and
list items (the section header and list entries such as "事实审查者", "高级工程师",
"安全专家", "一致性审查者", "冗余检查者") with no EOF newline; open the rules/zh/agents.md
content and add a single trailing newline character after the last list item so
the file ends with a newline.
In `@rules/zh/development-workflow.md`:
- Line 3: Update the Markdown link reference in the sentence containing
"[common/git-workflow.md](./git-workflow.md)" to use the correct link text
"[git-workflow.md](./git-workflow.md)"; locate the link in
rules/zh/development-workflow.md (the line that begins with "此文件扩展") and replace
the visible link label "common/git-workflow.md" with "git-workflow.md" while
keeping the target path "./git-workflow.md" unchanged.
---
Outside diff comments:
In `@CONTRIBUTING.md`:
- Around line 477-482: Update the "File Naming" section to explicitly exempt
skill files: clarify that the lowercase-with-hyphens rule (examples
`python-reviewer.md`, `tdd-workflow.md`) applies to general docs but does not
apply to files under the skills directory; add a short sentence noting the
skills convention is different — skills live in folders under skills/ and use
UPPERCASE `SKILL.md` inside a lowercase-hyphen skill folder (e.g.,
skills/my-skill/SKILL.md) so readers won’t interpret the global rule as
overriding the skills convention.
---
Nitpick comments:
In `@CONTRIBUTING.md`:
- Around line 148-153: The checklist item "Under 500 lines (800 max)" is
ambiguous—split it into two explicit rules so reviewers have a clear target and
a hard limit; replace that single line with two checklist entries such as
"Prefer target: under 500 lines" (or "Target: under 500 lines") and "Hard limit:
800 lines (do not exceed)", and ensure the wording appears in the checklist near
the other items so it reads as both a soft target and a firm maximum.
In `@rules/zh/agents.md`:
- Around line 20-27: The "立即使用代理" list omits the security-reviewer; update that
section (the "## 立即使用代理" block that currently lists planner, code-reviewer,
tdd-guide, and architect) to include security-reviewer as an immediate-use agent
(e.g., add a new bullet "5. 提交前进行安全分析 - 使用 **security-reviewer** 代理" or
similar), ensuring the agent name matches the agents table elsewhere so readers
can find its description.
In `@rules/zh/coding-style.md`:
- Line 48: 该文件在结尾缺少换行;在 rules/zh/coding-style.md 的末尾(可定位到最后一行包含文本 "[ ]
没有变更(使用不可变模式)")添加一个终止换行符(newline),确保文件以单个换行符结尾并保存改动以通过 POSIX 文件结尾检查。
In `@rules/zh/development-workflow.md`:
- Line 38: The file is missing a trailing newline; open the document containing
the line "参见 [git-workflow.md](./git-workflow.md) 了解提交消息格式和 PR 流程" and add a
single newline character at the end of the file so the file ends with a blank
line (ensure the EOF has a newline).
In `@rules/zh/git-workflow.md`:
- Line 24: The file's final line ("参见
[development-workflow.md](./development-workflow.md).") is missing a trailing
newline; open the markdown file and ensure it ends with a single newline
character (add an empty line or press Enter after that last line) so the file
terminates with a newline.
In `@rules/zh/hooks.md`:
- Line 30: The file "rules/zh/hooks.md" is missing a trailing newline after the
final line "误解的需求"; open the file and add a single newline character at the end
of the file so the file ends with a newline (ensure the final token "误解的需求" is
followed by a newline).
In `@rules/zh/patterns.md`:
- Line 50: The file patterns.md is missing a trailing newline; open the
rules/zh/patterns.md file (identify by the filename patterns.md) and add a
single newline character at the end of the file so it ends with '\n', then
save/commit the change.
- Line 31: The file ends without a trailing newline; open rules/zh/patterns.md
and add a single newline character at the end of the file so the last line
"包含分页响应的元数据(total、page、limit)" is terminated by a newline (ensure your editor or
git commit preserves the final newline).
In `@rules/zh/performance.md`:
- Line 55: The file ends without a newline after the final line "4. 每次修复后验证";
open the markdown file containing that line and add a single trailing newline
character at the end of the file (ensure the file ends with a blank line/line
break) so standard tools and POSIX conventions detect the EOF newline correctly.
In `@rules/zh/README.md`:
- Around line 99-102:
将说明文本中的措辞从“可能被语言特定文件覆盖”替换为更自然的“会被语言特定文件覆盖”以改善中文表达;在文件中查找包含该短语(例如在
rules/zh/README.md
中标记规则说明段落或引文区域)并直接替换该片段,保持其后续说明“:此规则可能被语言特定规则覆盖,对于某些语言,该模式可能不是惯用的。”相应一致(可同时将后半句中的“可能不是惯用的”保留或改为“可能不适用”以风格统一)。
In `@rules/zh/security.md`:
- Line 29: Add a trailing newline to the end of the file containing the line "5.
审查整个代码库中的类似问题" so the file ends with a single newline character (ensure the
final line is terminated); this fixes the missing EOF newline for
rules/zh/security.md.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 31aceba2-9dd2-45ed-9e44-a2cbb402dc88
📒 Files selected for processing (14)
CONTRIBUTING.mddocs/SKILL-DEVELOPMENT-GUIDE.mdrules/common/code-review.mdrules/zh/README.mdrules/zh/agents.mdrules/zh/code-review.mdrules/zh/coding-style.mdrules/zh/development-workflow.mdrules/zh/git-workflow.mdrules/zh/hooks.mdrules/zh/patterns.mdrules/zh/performance.mdrules/zh/security.mdrules/zh/testing.md
| @@ -0,0 +1,38 @@ | |||
| # 开发工作流 | |||
|
|
|||
| > 此文件扩展 [common/git-workflow.md](./git-workflow.md),包含 git 操作之前的完整功能开发流程。 | |||
There was a problem hiding this comment.
Correct the file path reference.
The reference to [common/git-workflow.md](./git-workflow.md) should be [git-workflow.md](./git-workflow.md) to match the actual file location in the same directory.
📝 Proposed fix
-> 此文件扩展 [common/git-workflow.md](./git-workflow.md),包含 git 操作之前的完整功能开发流程。
+> 此文件扩展 [git-workflow.md](./git-workflow.md),包含 git 操作之前的完整功能开发流程。🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@rules/zh/development-workflow.md` at line 3, Update the Markdown link
reference in the sentence containing
"[common/git-workflow.md](./git-workflow.md)" to use the correct link text
"[git-workflow.md](./git-workflow.md)"; locate the link in
rules/zh/development-workflow.md (the line that begins with "此文件扩展") and replace
the visible link label "common/git-workflow.md" with "git-workflow.md" while
keeping the target path "./git-workflow.md" unchanged.
|
thanks, queued for review. |
- Add missing code-review.md and development-workflow.md to zh/README.md directory listing - Add mkdir -p command before copy in manual install instructions - Fix TypeScript test command path in SKILL-DEVELOPMENT-GUIDE.md - Add Anti-Patterns section to SKILL.md template - Add Template category to Skill Categories table in CONTRIBUTING.md - Add Pre-Review Requirements section to code-review.md (both en and zh) - Add Pre-Review Checks step to development-workflow.md (both en and zh) - Add trailing newlines to all files that were missing them
|
Analysis Failed
Troubleshooting
Retry: |
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@rules/zh/README.md`:
- Around line 106-108: Change the awkward active phrasing in the sentence that
follows the `rules/common/` heading: replace "此规则可能被语言特定规则覆盖,对于某些语言,该模式可能不是惯用的。"
and the preceding fragment "规则会标记:" with a more natural passive form such as
"规则会被标记:" (and adjust the explanatory sentence if needed to match passive
voice). Locate the exact text under the `rules/common/` section in
rules/zh/README.md and update the wording to use "规则会被标记:" and any corresponding
sentence to passive voice so the Chinese reads naturally.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: 34bded1e-a6a1-4288-8b6a-6fc1c3901a45
📒 Files selected for processing (15)
CONTRIBUTING.mddocs/SKILL-DEVELOPMENT-GUIDE.mdrules/common/code-review.mdrules/common/development-workflow.mdrules/zh/README.mdrules/zh/agents.mdrules/zh/code-review.mdrules/zh/coding-style.mdrules/zh/development-workflow.mdrules/zh/git-workflow.mdrules/zh/hooks.mdrules/zh/patterns.mdrules/zh/performance.mdrules/zh/security.mdrules/zh/testing.md
✅ Files skipped from review due to trivial changes (12)
- rules/common/development-workflow.md
- rules/zh/security.md
- rules/zh/agents.md
- rules/zh/hooks.md
- rules/common/code-review.md
- rules/zh/coding-style.md
- rules/zh/git-workflow.md
- docs/SKILL-DEVELOPMENT-GUIDE.md
- CONTRIBUTING.md
- rules/zh/code-review.md
- rules/zh/performance.md
- rules/zh/patterns.md
🚧 Files skipped from review as they are similar to previous changes (2)
- rules/zh/testing.md
- rules/zh/development-workflow.md
There was a problem hiding this comment.
1 issue found across 15 files (changes from recent commits).
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="docs/SKILL-DEVELOPMENT-GUIDE.md">
<violation number="1" location="docs/SKILL-DEVELOPMENT-GUIDE.md:609">
P2: Validation instructions are internally inconsistent about working directory, causing documented commands to fail when run from inside a skill directory.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
|
Analysis Failed
Troubleshooting
Retry: |
Summary
This PR adds a comprehensive Skill Development Guide to help contributors create effective skills for Everything Claude Code.
Changes
New File:
docs/SKILL-DEVELOPMENT-GUIDE.mdA 900+ line comprehensive guide covering:
Updated:
CONTRIBUTING.mdEnhanced the "Contributing Skills" section:
Motivation
The project has 127 skills but lacked:
This guide fills that gap, making it easier for new contributors to create high-quality skills.
Testing
Type
Checklist
Summary by CodeRabbit