You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
|`code-review`| Apply team code quality standards |
34
+
|`code-checklist`| Apply team code quality standards |
35
35
36
36
37
37
@@ -56,7 +56,7 @@ Learn what skills are, why they matter, and how they differ from agents and MCP.
56
56
```
57
57
This shows all skills Copilot can find in your project and personal folders.
58
58
59
-
2.**Look at a real skill file:** Check out our provided [code-review SKILL.md](../.github/skills/code-review/SKILL.md) to see the pattern. It's just YAML frontmatter plus markdown instructions.
59
+
2.**Look at a real skill file:** Check out our provided [code-checklist SKILL.md](../.github/skills/code-checklist/SKILL.md) to see the pattern. It's just YAML frontmatter plus markdown instructions.
60
60
61
61
3.**Understand the core concept:** Skills are task-specific instructions that Copilot loads *automatically* when your prompt matches the skill's description. You don't need to activate them, just ask naturally.
62
62
@@ -68,16 +68,16 @@ Agent Skills are folders containing instructions, scripts, and resources that Co
68
68
```bash
69
69
copilot
70
70
71
-
>Review books.py for code quality issues
72
-
# Copilot detects this matches your "code-review" skill
71
+
>Check books.py against our quality checklist
72
+
# Copilot detects this matches your "code-checklist" skill
73
73
# and automatically applies its Python quality checklist
74
74
75
75
> Generate tests for the BookCollection class
76
76
# Copilot loads your "pytest-gen" skill
77
77
# and applies your preferred test structure
78
78
79
79
> What are the code quality issues in this file?
80
-
# Copilot loads your "code-review" skill
80
+
# Copilot loads your "code-checklist" skill
81
81
# and checks against your team's standards
82
82
```
83
83
@@ -92,15 +92,15 @@ While auto-triggering is the primary way skills work, you can also **invoke skil
92
92
```bash
93
93
> /generate-tests Create tests for the user authentication module
94
94
95
-
> /code-review Check books.py for code quality issues
95
+
> /code-checklist Check books.py for code quality issues
96
96
97
-
> /security-audit Review the API endpoints
97
+
> /security-audit Check the API endpointsfor vulnerabilities
98
98
```
99
99
100
100
This gives you explicit control when you want to ensure a specific skill is used.
101
101
102
102
> 📝 **Skills vs Agents Invocation**: Don't confuse skill invocation with agent invocation:
103
-
> -**Skills**: `/skill-name <prompt>`, e.g., `/code-review Check this file`
103
+
> -**Skills**: `/skill-name <prompt>`, e.g., `/code-checklist Check this file`
104
104
> -**Agents**: `/agent` (select from list) or `copilot --agent <name>` (command line)
105
105
>
106
106
> If you have both a skill and an agent with the same name (e.g., "code-reviewer"), typing `/code-reviewer` invokes the **skill**, not the agent.
@@ -129,7 +129,7 @@ Skills are just one piece of GitHub Copilot's extensibility model. Here's how th
|**MCP**| Connects external services | Need live data from APIs |
131
131
132
-
Use agents for broad expertise, skills for specific task instructions, and MCP for external data. An agent can use one or more skills during a conversation. For example, when you ask a code review agent to review your code, it might apply both a `security-audit` skill and a `style-guide` skill automatically.
132
+
Use agents for broad expertise, skills for specific task instructions, and MCP for external data. An agent can use one or more skills during a conversation. For example, when you ask an agent to check your code, it might apply both a `security-audit` skill and a `code-checklist` skill automatically.
133
133
134
134
> 📚 **Learn More**: See the official [About Agent Skills](https://docs.github.com/copilot/concepts/agents/about-agent-skills) documentation for the complete reference on skill formats and best practices.
135
135
@@ -162,17 +162,17 @@ Time: **30+ seconds** to type. Consistency: **varies by memory**.
162
162
163
163
### After Skills: Automatic Best Practices
164
164
165
-
With a `code-review` skill installed, just ask naturally:
165
+
With a `code-checklist` skill installed, just ask naturally:
166
166
167
167
```bash
168
168
copilot
169
169
170
-
>Review the book collection code for issues
170
+
>Check the book collection code for quality issues
171
171
```
172
172
173
173
**What happens behind the scenes**:
174
-
1. Copilot sees "review" and "issues" in your prompt
175
-
2. Checks skill descriptions, finds your `code-review` skill matches
174
+
1. Copilot sees "code quality" and "issues" in your prompt
175
+
2. Checks skill descriptions, finds your `code-checklist` skill matches
176
176
3. Automatically loads your team's quality checklist
177
177
4. Applies all checks without you listing them
178
178
@@ -182,7 +182,7 @@ copilot
182
182
183
183
**Output**:
184
184
```
185
-
## Code Review: books.py
185
+
## Code Checklist: books.py
186
186
187
187
### Code Quality
188
188
- [PASS] All functions have type hints
@@ -300,14 +300,14 @@ Skills use a simple markdown format with YAML frontmatter:
300
300
301
301
```markdown
302
302
---
303
-
name: code-review
304
-
description: Comprehensive code review with security, performance, and maintainability checks
303
+
name: code-checklist
304
+
description: Comprehensive code quality checklist with security, performance, and maintainability checks
305
305
license: MIT
306
306
---
307
307
308
-
# Code Review
308
+
# Code Checklist
309
309
310
-
When reviewing code, check for:
310
+
When checking code, look for:
311
311
312
312
## Security
313
313
- SQL injection vulnerabilities
@@ -445,9 +445,9 @@ Skills and agents work together. The agent provides expertise, the skill provide
445
445
# Start with a code-reviewer agent
446
446
copilot --agent code-reviewer
447
447
448
-
>Review the book app for issues
448
+
>Check the book app for quality issues
449
449
# code-reviewer agent's expertise combines
450
-
# with your code-review skill's checklist
450
+
# with your code-checklist skill's checklist
451
451
```
452
452
453
453
---
@@ -484,7 +484,7 @@ copilot
484
484
Available skills:
485
485
- security-audit: Security-focused code review checking OWASP Top 10
486
486
- generate-tests: Generate comprehensive unit tests with edge cases
> ⚠️ **Review before installing**: Always read a skill's `SKILL.md` before copying it into your project. Skills control what Copilot does, and a malicious skill could instruct it to run harmful commands or modify code in unexpected ways.
@@ -812,7 +812,7 @@ Run `/skills reload` after creating or editing skills to ensure changes are pick
812
812
813
813
**Testing if a skill loads** - Ask Copilot directly:
814
814
```bash
815
-
> What skills do you have available for code review?
815
+
> What skills do you have available forchecking code quality?
0 commit comments