x-ipe-tool-task-board-manager

name	x-ipe-tool-task-board-manager
description	Manage tasks and features on the JSON-based project boards. Use for creating, updating, querying, and archiving tasks and features. Triggers on requests like "create task", "update task status", "query tasks", "create feature", "update feature", "query features".

Task & Feature Board Manager

Purpose

Provides CRUD operations for the JSON-based task board and feature board. All project tracking data is stored as JSON files managed by Python scripts with atomic file operations and concurrency-safe locking.

Data Storage

Board	Location	Format
Tasks	x-ipe-docs/planning/tasks/	Daily files tasks-YYYY-MM-DD.json + tasks-index.json
Features	x-ipe-docs/planning/features/features.json	Single file with all features

Available Scripts

All scripts are in .github/skills/x-ipe-tool-task-board-manager/scripts/.

Task Operations

Script	Purpose	Example
task_create.py	Create a new task	uv run python scripts/task_create.py --title "Fix bug" --description "..." --assignee "Drift"
task_update.py	Update task fields	uv run python scripts/task_update.py TASK-123 --status done
task_query.py	Query/search tasks	uv run python scripts/task_query.py --status in_progress
task_archive.py	Archive a task	uv run python scripts/task_archive.py TASK-123

Feature Operations

Script	Purpose	Example
feature_create.py	Create a new feature	uv run python scripts/feature_create.py --epic-id EPIC-055 --feature-id FEATURE-055-A --title "..." --status Planned
feature_update.py	Update feature fields	uv run python scripts/feature_update.py FEATURE-055-A --status Completed
feature_query.py	Query/search features	uv run python scripts/feature_query.py --epic-id EPIC-055

Migration

Script	Purpose	Example
migrate.py	Migrate from markdown boards to JSON	uv run python scripts/migrate.py --all --dry-run

Shared Library

_board_lib.py provides the foundation:

• resolve_data_path(board_type) — Locates data directory
• atomic_read_json(path) / atomic_write_json(path, data) — Thread-safe I/O
• with_file_lock(path) — fcntl-based file locking
• generate_task_id() / validate_task(data) / validate_feature(data) — ID generation & schema validation

Task Data Model

task_id: "TASK-XXX"        # Auto-generated sequential ID
title: "Task title"
description: "Detailed description"
status: "pending | in_progress | done | blocked | cancelled"
assignee: "Agent nickname"
created_at: "YYYY-MM-DD"
updated_at: "YYYY-MM-DD"
output_links: []            # Paths to artifacts produced
next_task: ""               # Follow-up task reference
_version: "1.0"

Feature Data Model

feature_id: "FEATURE-XXX-Y"
epic_id: "EPIC-XXX"
title: "Feature title"
status: "Planned | Refined | Designed | Implemented | Tested | Completed | Retired"
version: "1.0"
description: ""
dependencies: []
artifact_links: []
_version: "1.0"

Usage Pattern (for other skills)

# Creating a task
Use x-ipe-tool-task-board-manager to create a task:
  uv run python .github/skills/x-ipe-tool-task-board-manager/scripts/task_create.py \
    --title "TASK_TITLE" --description "TASK_DESC" --assignee "NICKNAME"

# Updating task status
Use x-ipe-tool-task-board-manager to update task:
  uv run python .github/skills/x-ipe-tool-task-board-manager/scripts/task_update.py \
    TASK-XXX --status done

# Querying tasks
Use x-ipe-tool-task-board-manager to query tasks:
  uv run python .github/skills/x-ipe-tool-task-board-manager/scripts/task_query.py \
    --status in_progress --assignee "NICKNAME"

Concurrency

• All operations use fcntl.flock file locking
• Tasks: nested locks (daily file → index)
• Features: single lock on features.json
• NEVER make parallel write calls — they must be sequential

Notes

• No deletion: tasks are archived via *.archived.json; features use Retired status
• Schema validation rejects unknown fields
• All scripts output JSON to stdout for programmatic consumption