AgentSkillsCN

backlog-management

基于CSV系统，对项目待办事项进行查询、更新与管理。

SKILL.md
--- frontmatter
name: backlog-management
description: Query, update, and manage the project backlog using the CSV-based system.

Backlog Management Skill

This skill provides workflows for managing the project backlog. The backlog uses a CSV-based system for efficient querying and reduced token usage.

System Overview

code
.claude/plans/backlog/
├── data/
│   ├── backlog.csv      # Main table (source of truth)
│   ├── sprints.csv      # Sprint history
│   ├── changelog.csv    # Audit trail
│   └── SCHEMA.md        # Column definitions
├── scripts/
│   ├── queries.py       # Query interface
│   └── validate.py      # Schema validation
├── items/               # BACKLOG-XXX.md detail files
└── README.md            # Quick start guide

Quick Reference

Query Items

bash
# By status
python .claude/plans/backlog/scripts/queries.py status pending

# By priority
python .claude/plans/backlog/scripts/queries.py priority high --status pending

# Sprint items
python .claude/plans/backlog/scripts/queries.py sprint SPRINT-042

# Search
python .claude/plans/backlog/scripts/queries.py search "sync"

# Statistics
python .claude/plans/backlog/scripts/queries.py stats

Python Module Usage

python
import csv

# Load all items
with open('.claude/plans/backlog/data/backlog.csv') as f:
    items = list(csv.DictReader(f))

# Filter to pending high priority
pending_high = [
    i for i in items
    if i['status'].lower() == 'pending'
    and i['priority'].lower() == 'high'
]

Workflows

WorkflowWhen to Use
Backlog AnalysisGenerate health report, find attention items
Add ItemCreating a new backlog item
Close ItemCompleting or obsoleting an item
Sprint PlanningPlanning a new sprint

Key Rules

  1. •CSV is source of truth - Always update backlog.csv for status changes
  2. •All items need .md files - Create BACKLOG-XXX.md for every item
  3. •Run validation - Execute python .claude/plans/backlog/scripts/validate.py before committing
  4. •Log key changes - Add entries to changelog.csv for significant changes
  5. •Column order matters - The CSV column order is: id,title,category,priority,status,sprint,est_tokens,actual_tokens,variance,created_at,completed_at,file

CSV Column Order (Quick Reference)

code
id,title,category,priority,status,sprint,est_tokens,actual_tokens,variance,created_at,completed_at,file

Common Mistake: Swapping category/priority/status positions. Always verify positions 3-5:

  • •Position 3: category (enhancement, bug, feature, etc.)
  • •Position 4: priority (high, medium, low, critical)
  • •Position 5: status (pending, in-progress, testing, etc.)

Status Flow (IMPORTANT)

code
pending → in-progress → testing → completed
                           ↓
                       reopened → in-progress → ...

CRITICAL RULES:

  1. •Code merged = testing (NOT completed)
  2. •Only user verification = completed
  3. •Failed testing = reopened (NEVER create new task)

Status Values

  • •pending - Not started
  • •in-progress - Currently being worked on
  • •testing - Code merged, awaiting user verification
  • •completed - Done AND verified by user
  • •blocked - Waiting on something
  • •deferred - Postponed
  • •obsolete - No longer relevant
  • •reopened - Failed testing, needs more work

Priority

  • •critical - Must be done immediately
  • •high - Important, do soon
  • •medium - Normal priority
  • •low - Nice to have

Related Documentation