Nettoyage en lot des titres de tâches avec MCP : nommage cohérent en quelques minutes
Standardize task titles across your workspace using MCP. This batch editing playbook shows you how to list tasks matching a pattern, propose standardized titles, show a diff preview, and apply updates only after explicit approval.
Ce que ce workflow permet
Batch title cleanup with MCP standardizes task naming:
Résultats clés
- Pattern matching: Find tasks with inconsistent naming patterns
- Standardization: Propose consistent title formats
- Diff preview: Show before/after comparison before changes
- Batch updates: Apply standardized titles to multiple tasks
- Approval workflow: Require explicit approval before any updates
Prérequis
Before using this workflow, ensure you have:
- Corcava MCP server configured in your AI assistant
- API key with read access to tasks
- Tasks with inconsistent naming that need standardization
- For batch updates: API key with write permissions
Workflow pas à pas
Step 1: Find Tasks Matching Pattern
Start by identifying tasks that need cleanup:
Pattern Matching Prompt
What the AI does:
- Calls
list_tasksto get tasks - Analyzes task titles for patterns
- Identifies tasks matching criteria
- Presents list of tasks needing cleanup
Step 2: Propose Standardized Titles
Generate consistent title formats:
Standardization Prompt
What the AI does:
- Analyzes each task title
- Applies standardization rules
- Generates proposed new titles
- Presents before/after comparison
Step 3: Show Diff Preview
Display a clear before/after comparison:
Diff Preview Prompt
What the AI does:
- Formats changes as clear diff
- Groups changes logically
- Highlights modifications
- Presents reviewable diff
Step 4: Get Approval
⚠️ Always Require Explicit Approval
When applying batch updates:
- Show Diff First: Display all changes before applying
- Approval Format: Provide copyable approval format
- Confirmation Token: Require explicit "APPROVE ALL" or similar
- Review Period: Give user time to review changes
Approval Request Prompt
What the AI does:
- Shows diff preview
- Provides copyable approval format
- Waits for "APPROVE ALL" or modifications
- Only proceeds after explicit approval
Step 5: Apply Updates (After Approval)
Update tasks only after approval:
Batch Update Prompt
What the AI does:
- Verifies approval token
- Calls
update_taskfor each approved change - Reports progress and results
- Confirms successful updates
Complete Workflow Prompts
Here are complete, copy-paste ready prompts:
Full Batch Cleanup Workflow
Use this when: You want a complete batch cleanup workflow
Quick Title Standardization
Use this when: You want to standardize titles for a specific project
Workflow Variants
Variant 1: Project Prefix Standardization
Add consistent project prefixes:
What happens: AI adds consistent project prefixes to task titles
Variant 2: Capitalization Fix
Standardize capitalization:
What happens: AI standardizes capitalization across tasks
Variant 3: Format Standardization
Apply consistent title format:
What happens: AI applies consistent format template to all tasks
Example: Complete Batch Cleanup Session
Here's a real example of how a batch cleanup session flows:
User Prompt
AI Response (Diff Preview)
Proposed Title Changes (5 tasks):
- Current: "implement auth endpoints" → Proposed: "[API] Implement: Auth Endpoints"
- Current: "Write tests" → Proposed: "[API] Write: Tests"
- Current: "API: Document endpoints" → Proposed: "[API] Document: Endpoints"
- Current: "fix bug in login" → Proposed: "[API] Fix: Bug in Login"
- Current: "Update API docs" → Proposed: "[API] Update: API Docs"
Review the changes above. If you approve all changes, type: APPROVE ALL
Safety Patterns
⚠️ Approval Rules for Batch Updates
When applying batch title updates:
- Always show diff: Display all changes before applying
- Require explicit approval: Use "APPROVE ALL" token for batch operations
- Provide copyable format: Make approval easy to copy/paste
- Allow modifications: Let user modify specific titles before approval
- Report results: Confirm each update was successful
Troubleshooting
Too Many Tasks Found
Symptom: Pattern matches too many tasks
Possible causes:
- Pattern is too broad
- Many tasks need cleanup
Fix: Narrow scope: "Find tasks in project [X] with pattern [Y]" or limit to specific date range
Proposed Titles Not Good
Symptom: Standardized titles don't match expectations
Possible causes:
- Format specification was unclear
- AI misunderstood the pattern
Fix: Provide examples: "Use this format: '[Project] Verb: Description'. Example: '[API] Implement: Auth Endpoints'"
Related Tools
This workflow uses these Corcava MCP tools:
list_tasks
Find tasks matching patterns
update_task
Apply standardized title updates
get_task
Get task details for analysis
Related Use Cases
Bug Triage
Standardize bug task titles during triage
Backlog Grooming
Clean up titles while grooming backlog
Standardize Task Titles Automatically
Connect your AI assistant to Corcava and clean up task naming in minutes