shared-space

name	shared-space
description	Manages shared ledgers for family or team collaboration, including shared account summaries, spending statistics, and member settlements. Use when user asks about shared expenses, family ledger, team budget, or joint accounts. 关键词：共享账本、家庭账单、团队支出、AA制。
license	Apache-2.0
metadata	{"type":"collaboration","version":"1.1"}
allowed-tools	execute read_file ls

Skill: Shared Space

Responsible for handling collaboration logic for family or team shared ledgers.

Use Cases

• User asks: "What is my family's spending this month?"
• User requests: "Show me the team expense summary."
• User asks: "How much did we spend together?"

Available Scripts

Script	Purpose	Output
app/skills/shared-space/scripts/list_spaces.py	List user's shared spaces	Space list JSON
app/skills/shared-space/scripts/query_space_summary.py	Query spending summary	Summary with monthly stats

list_spaces.py - Get Available Spaces

uv run python app/skills/shared-space/scripts/list_spaces.py

Returns a list of shared spaces the user has access to, with role information.

query_space_summary.py - Space Summary

uv run python app/skills/shared-space/scripts/query_space_summary.py

Query a specific space:

echo '{"space_id": "uuid-string"}' | uv run python app/skills/shared-space/scripts/query_space_summary.py

Returns monthly spending statistics for the space.

Workflows

1. Discovery - List Available Spaces

Triggers: "What shared spaces do I have?", "Show my shared accounts"

Run list_spaces.py and present the available spaces to the user.

2. Summary Query

Triggers: "Family spending this month", "Team expense report"

Run query_space_summary.py and summarize the spending data.

Rules

• Only used for processing ledger data with "shared" or "multi-user" context.
• For transfers, guide the user to use the Transfer Expert skill.
• For recording transactions to a space, use the normal record_transactions tool and the AI should naturally associate it if context indicates a shared space.
• Localization: Localize all your responses back to the current session language.