MCP Examples
Real-world examples of how to use Erold with AI assistants via MCP.
Task Management
Creating a Task
You: "Create a task in MYAPP to implement user authentication with JWT"
Claude:
I'll create that task for you.
[Uses erold_task_create with:
projectId: "proj_xxx"
title: "Implement user authentication with JWT"
priority: "high"
labels: ["feature", "auth"]]
✓ Created task MYAPP-47: "Implement user authentication with JWT"
Priority: High
Labels: feature, auth Updating Task Status
You: "Mark MYAPP-47 as in progress"
Claude:
[Uses erold_task_update with:
taskId: "MYAPP-47"
status: "in_progress"]
✓ Updated MYAPP-47 status to "In Progress" Finding Tasks
You: "What are my high priority tasks?"
Claude:
[Uses erold_task_list with:
status: "todo"
priority: "high"]
Found 3 high-priority tasks:
1. MYAPP-47: Implement user authentication with JWT
2. MYAPP-45: Fix production database timeout
3. MYAPP-42: Add rate limiting to API Knowledge Base
Storing Project Context
You: "Save to knowledge base: Our API uses REST with JSON. Auth is via Bearer tokens with 24h expiry. Rate limit is 100 req/min."
Claude:
[Uses erold_knowledge_set with:
key: "api/overview"
content: "Our API uses REST with JSON..."]
✓ Saved to knowledge base with key "api/overview" Retrieving Context
You: "What's our API authentication approach?"
Claude:
[Uses erold_knowledge_search with:
query: "API authentication"]
Based on your knowledge base:
Your API uses Bearer token authentication with 24-hour expiry.
Rate limit is 100 requests per minute. Project Overview
Getting Project Status
You: "Give me an overview of the MYAPP project"
Claude:
[Uses erold_project_get with projectId]
[Uses erold_task_list with projectId]
MYAPP - My Application
Statistics:
- Total tasks: 47
- Todo: 12
- In Progress: 8
- Done: 27
Recent activity:
- MYAPP-47 created (2 hours ago)
- MYAPP-46 completed (yesterday)
- MYAPP-45 assigned to John (yesterday) Daily Workflow
Morning Standup
You: "What did I work on yesterday and what's planned for today?"
Claude:
[Uses erold_activity_list]
[Uses erold_task_list with status: "in_progress"]
Yesterday:
- Completed MYAPP-44: Add email validation
- Started MYAPP-45: Fix database timeout
- Added 3 comments on MYAPP-45
Today (In Progress):
- MYAPP-45: Fix database timeout (High priority)
- MYAPP-46: Update user dashboard
Blocked:
- None currently End of Day Update
You: "I finished the database timeout fix. Log 3 hours and mark it done."
Claude:
[Uses erold_time_log with:
taskId: "MYAPP-45"
minutes: 180
description: "Fixed timeout issue"]
[Uses erold_task_update with:
taskId: "MYAPP-45"
status: "done"]
✓ Logged 3 hours on MYAPP-45
✓ Marked MYAPP-45 as Done Code Review Context
Understanding a Task Before Review
You: "I'm about to review PR #123 for MYAPP-45. What's the context?"
Claude:
[Uses erold_task_get with taskId: "MYAPP-45"]
[Uses erold_comment_list with taskId: "MYAPP-45"]
[Uses erold_knowledge_search with query: "database"]
MYAPP-45: Fix production database timeout
Description:
Database queries timing out under load. Need to add
connection pooling and optimize slow queries.
Comments:
- John: "Main issue is the user list query"
- Sarah: "Also check the report generation"
Related knowledge:
- Database uses PostgreSQL 14
- Connection limit is 100
- Current pool size: 10 Best Practices
- Be specific - Include project names and task IDs when possible
- Use knowledge base - Store context AI should know across sessions
- Keep tasks updated - Ask AI to update status as you work
- Log time - Track time for better project insights