@moon791017/neo-skills 1.1.17 → 1.1.19

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -85,7 +85,36 @@ BREAKING CHANGE: 當發現項目等於或高於閾值時,掃描現在會使 CI
85
85
  `;
86
86
 
87
87
  export const factCheckInstructions = `
88
- You must conduct "fact-check thinking" before answering. Do not assume, speculate, or create content unless explicitly provided by the user or clearly existing in the data. Strictly adhere to sources: only use content provided by the user, explicitly documented knowledge within your system, or clearly verified information. If information is insufficient, state directly "insufficient data" or "I cannot be certain"—do not speculate. Show the basis of your thinking: if you cite data or make inferences, explain the section or reason you rely on. If it is personal analysis or estimation, it must be explicitly labeled as "this is an inference" or "this is a hypothetical scenario". Avoid pretending to know: do not "complete" non-existent content to make the answer complete. If encountering ambiguous or incomplete questions, ask for clarification or present options first rather than deciding on your own. Maintain semantic consistency: do not rewrite or expand the user's original intent. If you need to restate, it must be explicitly labeled as "restated version" and maintain semantic equivalence. Answer format: if there is clear data, answer and attach the basis. If there is no clear data, answer "cannot be certain" and explain why. Do not use ambiguous tones such as "should be", "probably", "I guess" in your answers unless requested by the user. Depth of thinking: before generating output, check whether the answer: a. has a clear basis, b. does not exceed the scope of the question, c. does not contain any names, numbers, events, or assumptions not explicitly mentioned. Ultimate principle: Better to leave it blank than to fabricate.`;
88
+ ## Fact-Check Thinking Guidelines
89
+
90
+ Conduct strict fact-checking before responding. Never speculate, assume, or fabricate missing details.
91
+
92
+ ### Core Rules
93
+ 1. **Strict Source Grounding**: Rely strictly on provided context, explicit documentation, or verified facts. If data is missing, reply directly with "insufficient data" or "I cannot be certain".
94
+ 2. **Explicit Basis & Labels**: Cite source sections for facts. Explicitly mark personal analysis or estimations as "inference" or "hypothetical scenario".
95
+ 3. **Preserve Intent & Clarify**: Do not expand or mutate user intent. For ambiguous requests, ask clarifying questions or present options instead of guessing.
96
+ 4. **No Speculative Tone**: Avoid uncertain phrasing such as "should be", "probably", or "I guess".
97
+ 5. **Validation Check**: Verify that output has a clear source basis and contains no unmentioned names, figures, or assumptions.
98
+
99
+ **Ultimate Rule**: Better to leave unstated than to fabricate.`;
100
+
101
+ export const minimalOutputInstructions = `
102
+ ## Minimal Output Guidelines
103
+
104
+ Run all build, test, deploy, and formatting commands with minimal output flags (quiet, silent, summary-only) to preserve context window capacity.
105
+
106
+ ### Rules
107
+ 1. **Use Quiet Mode by Default**: Prefer \`-q\`, \`--quiet\`, \`--silent\`, or \`--loglevel=error\` for routine CLI runs.
108
+ 2. **Verbose Only on Error**: Enable \`--verbose\` or debug logging only when diagnosing a failed command.
109
+ 3. **Summarize Tests**: Suppress passing test lists and progress bars; output failure tracebacks and summary counts only.
110
+
111
+ ### Recommended Flags
112
+ - **Package Managers**: \`npm install --quiet\`, \`pnpm install --silent\`, \`yarn install --silent\`
113
+ - **Test Runners**: \`vitest run --reporter=basic\`, \`jest --silent\`, \`pytest -q --tb=short\`, \`dotnet test -v q --nologo\`, \`cargo test -- --quiet\`
114
+ - **Build**: \`dotnet build -v q --nologo\`, \`cargo build --quiet\`, \`go build\`
115
+ - **Lint & Format**: \`eslint --quiet\`, \`prettier --check --loglevel warn\`
116
+ - **Deploy & Infra**: \`docker build -q .\`, \`terraform plan -no-color\`
117
+ `;
89
118
 
90
119
  /**
91
120
  * 系統提示詞 Registry — 以 kebab-case 作為 key,供 CLI 選擇安裝。
@@ -107,4 +136,8 @@ export const INSTRUCTIONS = {
107
136
  name: 'Fact-Check Thinking',
108
137
  content: factCheckInstructions,
109
138
  },
139
+ 'minimal-output': {
140
+ name: 'Minimal Output Commands',
141
+ content: minimalOutputInstructions,
142
+ },
110
143
  };
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@moon791017/neo-skills",
3
- "version": "1.1.17",
3
+ "version": "1.1.19",
4
4
  "type": "module",
5
5
  "description": "Neo Skills: A Universal AI Agent Skills Extension",
6
6
  "homepage": "https://neo-blog-iota.vercel.app/",
@@ -85,8 +85,9 @@ Construct the PR Title and Description based on extracted commit and diff contex
85
85
  - Keep under 50 characters, imperative mood, direct and actionable.
86
86
 
87
87
  #### 2. PR Output Standard Structure
88
- The PR output must strictly consist of only the following two sections:
88
+ The PR output **must strictly be wrapped in a Markdown code block** (i.e. raw Markdown source code ` ```markdown ... ``` `) so that users can easily copy it directly into GitHub/GitLab PR description fields. The content inside the code block must consist exclusively of the following two sections:
89
89
 
90
+ ````markdown
90
91
  ```markdown
91
92
  標題:
92
93
  <Conventional Commits 格式標題>
@@ -94,6 +95,7 @@ The PR output must strictly consist of only the following two sections:
94
95
  內容:
95
96
  - 列點並重點化的說明重要變更內容(包含主要模組/檔案調整、邏輯變更與重大影響)。
96
97
  ```
98
+ ````
97
99
 
98
100
  #### 3. Strict De-slop Rules
99
101
  * ❌ Never use: "It is worth noting that", "Needless to say", "This PR aims to", "In summary", "Clearly".
@@ -4,8 +4,9 @@
4
4
  {
5
5
  "id": 1,
6
6
  "prompt": "Generate a concise, non-AI-slop PR title and description for merging feature/user-auth into main.",
7
- "expected_output": "A PR title adhering to Conventional Commits format under '標題:', and a bulleted markdown PR description under '內容:' with no AI fluff phrases.",
7
+ "expected_output": "A PR title adhering to Conventional Commits format under '標題:', and a bulleted markdown PR description under '內容:' wrapped in a Markdown code block for easy copying, with no AI fluff phrases.",
8
8
  "assertions": [
9
+ "PR output is enclosed inside a Markdown code block for easy copying",
9
10
  "PR Title complies with Conventional Commits format under 標題:",
10
11
  "PR Description includes bullet points summarizing key changes under 內容:",
11
12
  "Content is structured strictly with bullet points",
@@ -23,8 +23,9 @@ Titles must be clear, concise, and directly reflect the underlying changes using
23
23
 
24
24
  ## 2. PR Description Template
25
25
 
26
- The PR output format must strictly follow:
26
+ The PR output **must strictly be wrapped in a Markdown code block** (i.e. raw Markdown source code ` ```markdown ... ``` `) so that users can easily copy it directly into GitHub/GitLab PR description fields. The output format must strictly follow:
27
27
 
28
+ ````markdown
28
29
  ```markdown
29
30
  標題:
30
31
  <type>(<scope>): <short summary>
@@ -33,6 +34,7 @@ The PR output format must strictly follow:
33
34
  - **[重要變更/關鍵模組 1]**:說明具體調整內容與目的。
34
35
  - **[重要變更/關鍵模組 2]**:說明具體調整內容與目的。
35
36
  ```
37
+ ````
36
38
 
37
39
  ---
38
40