# MCP Prompt Library — Ready-to-Use Templates

> AI Starter Package (aistarterpackage.com)
> Format inspired by Zack Deris's Apollo Prompt Library

## How to Use These Prompts

1. Copy the template
2. Replace anything in [BRACKETS] with your specific details
3. Paste into Claude Code with the relevant MCP server active
4. Claude executes the action

---

## GitHub MCP

### Prompt 1: Summarize All Open PRs

**When to use:** Morning review — get a quick status of all open pull requests.

**Template:**
```
List all open pull requests in [REPO_OWNER]/[REPO_NAME]. For each PR show:
- PR number and title
- Author
- How many days it has been open
- Number of comments
- Whether CI is passing or failing
Sort by oldest first.
```

**Example:**
```
List all open pull requests in aistarterpackage/jaime. For each PR show:
- PR number and title
- Author
- How many days it has been open
- Number of comments
- Whether CI is passing or failing
Sort by oldest first.
```

**Notes:**
- Free action, no API costs
- Works with any public or authorized private repo

---

### Prompt 2: Create an Issue from a Bug Description

**When to use:** When you find a bug and want to log it without leaving your terminal.

**Template:**
```
Create a new issue in [REPO_OWNER]/[REPO_NAME] with:
Title: [BUG_TITLE]
Body: ## Description
[DESCRIBE_THE_BUG]

## Steps to Reproduce
1. [STEP_1]
2. [STEP_2]
3. [STEP_3]

## Expected Behavior
[WHAT_SHOULD_HAPPEN]

## Actual Behavior
[WHAT_ACTUALLY_HAPPENS]

Labels: bug
```

---

### Prompt 3: Search Codebase for a Pattern

**When to use:** Find all occurrences of a pattern across the repo without cloning it locally.

**Template:**
```
Search [REPO_OWNER]/[REPO_NAME] for all files containing "[SEARCH_PATTERN]".
Show the file path, line number, and surrounding context for each match.
Group results by directory.
```

---

## Apify MCP

### Prompt 4: Scrape Google Results for Competitive Analysis

**When to use:** Research competitors or market landscape for a keyword.

**Template:**
```
Use Apify's Google Search Scraper to find the top [NUMBER] results for "[KEYWORD]".
For each result, extract:
- Title
- URL
- Meta description
- Position
Then analyze: who are the main competitors? What content formats dominate? What gaps exist?
```

**Notes:**
- Uses Apify credits
- Great for SEO research and content planning

---

### Prompt 5: Crawl a Website for RAG

**When to use:** Build a knowledge base from a documentation site or wiki.

**Template:**
```
Use Apify's Website Content Crawler to crawl [WEBSITE_URL].
Extract all pages as clean markdown.
For each page, return:
- Page URL
- Page title
- Full text content (no HTML tags)
Limit to [NUMBER] pages maximum.
```

---

## SerpAPI MCP

### Prompt 6: Multi-Engine Research

**When to use:** Comprehensive research on a topic across multiple search engines.

**Template:**
```
Search for "[TOPIC]" across Google, Google Scholar, and Google News.
For each source, give me the top 5 results with titles, URLs, and key takeaways.
Then synthesize: what are the main themes? Any conflicting viewpoints? What is the latest development?
```

---

### Prompt 7: Content Gap Analysis

**When to use:** Find questions people are asking that you haven't answered yet.

**Template:**
```
Search Google for "[YOUR_TOPIC]" and extract all "People Also Ask" questions.
Then search for "[YOUR_BRAND_NAME] [YOUR_TOPIC]" and compare.
Which questions do we already answer? Which ones are gaps we should create content for?
Prioritize by estimated search volume (high/medium/low).
```

---

## Slack MCP

### Prompt 8: Daily Standup Summary

**When to use:** Automate your morning standup message.

**Template:**
```
Send a message to the [CHANNEL_NAME] channel with this standup update:

*Daily Standup — [TODAY'S DATE]*

✅ *Yesterday:*
- [WHAT_YOU_DID_YESTERDAY]

🎯 *Today:*
- [WHAT_YOU_PLAN_TODAY]

🚧 *Blockers:*
- [ANY_BLOCKERS_OR_NONE]
```

---

### Prompt 9: Search Decision History

**When to use:** Find what was decided about a topic in past conversations.

**Template:**
```
Search our Slack message history for discussions about "[TOPIC]".
Find messages from the last [NUMBER] days that contain decisions, action items, or conclusions.
Summarize: What was decided? Who decided it? When? Are there any open questions?
```

---

## Browserbase MCP

### Prompt 10: Screenshot and Describe a Page

**When to use:** QA check or competitive research without opening a browser.

**Template:**
```
Navigate to [URL] using Browserbase.
Take a full-page screenshot.
Describe what you see:
- What is the page about?
- What is the primary CTA?
- What is the color scheme?
- How is the content structured?
- Any obvious UX issues?
```

---

## Pro Tips

### Chain Prompts Together
Claude remembers your conversation. After searching, say "enrich the top 3 from that list" — no need to repeat details.

### Free vs Paid Actions
- **Free:** GitHub search, GitHub issues, Slack messages, SerpAPI searches
- **Uses credits:** Apify scraping, contact enrichment, company data

### Best Practice
Always **search first, then act**. Search is usually free. Only spend credits on results worth pursuing.

---

*Get the full prompt library for all 18 MCP servers at aistarterpackage.com/ebooks*
*Format inspired by Zack Deris's Apollo Prompt Library — zackderis.com*