agent-budget
Budget management and cost tracking MCP server for autonomous agents, enabling budget creation, cost recording, spending projections, and alert rules.
README
Agent Budget
MCP server + CLI for autonomous agents to manage budgets, track expenses, control spending, set savings goals, and enforce spending rules.
Features
v0.2.0
- Savings Goals — Track progress toward savings targets with auto-completion
- Budget Rollover — Carry unspent budget forward to the next period
- Spending Rules — Block, warn, or require approval for expenses
- Expense Updates — Modify existing expenses (amount, vendor, receipt, status)
- Expense Receipts — Attach receipt URLs, mark expenses as reimbursable
- Alert Threshold Updates — Customize budget alert thresholds
- Deprecation Fixes —
datetime.utcnow()→datetime.now(timezone.utc)
v0.1.0
- Budget Management — Create, update, delete budgets with spending limits
- Expense Tracking — Log expenses with categories, tags, and vendor info
- Recurring Expenses — Schedule recurring payments (daily/weekly/monthly/quarterly/yearly)
- Budget vs. Actual — Compare spending against budget limits
- Spending Forecasts — Project future spending based on history
- Alert System — Automatic alerts at configurable thresholds
- Multi-Currency — Support for 15+ currencies
- Data Export — Export to JSON, CSV, or Markdown
Installation
pip install agent-budget
Or with uv:
uv pip install agent-budget
Quick Start
CLI
# Create a monthly budget
agent-budget budget create "API Costs" --limit 500 --period monthly --category api
# Create a budget with rollover
agent-budget budget create "Infrastructure" --limit 1000 --period monthly --rollover --rollover-cap 200
# Log an expense
agent-budget expense add 25.50 --category api --description "OpenAI GPT-4 call" --vendor "OpenAI"
# Log a reimbursable expense with receipt
agent-budget expense add 99.00 --category saas --vendor "GitHub" --reimbursable --receipt-url "https://receipts.example.com/gh-001"
# Update an expense
agent-budget expense update EXP-ABC12345 --amount 125.00 --vendor "AWS"
# Check budget status
agent-budget budget status
# Process budget rollovers
agent-budget budget rollover
# Create a savings goal
agent-budget savings create "Emergency Fund" --target 10000 --target-date 2027-01-01
# Contribute to a savings goal
agent-budget savings contribute SAV-ABC12345 --amount 500 --note "Monthly deposit"
# Withdraw from a savings goal
agent-budget savings withdraw SAV-ABC12345 --amount 200 --note "Emergency repair"
# Add a spending rule
agent-budget rule add "API Cap" --category api --action block --threshold-amount 500
# Add an approval rule
agent-budget rule add "Large Expenses" --category infra --action block --approval-above 100
# Check if an expense would violate rules
agent-budget rule check --amount 150 --category infra
# Set up a recurring expense
agent-budget recurring add "AWS Hosting" --amount 99 --category infra --frequency monthly
# Get spending summary
agent-budget summary --this-month
# Get spending forecast
agent-budget forecast --months 3
# Check alerts
agent-budget alerts
# Export data
agent-budget export --format json
MCP Server
Start the MCP server for integration with AI agents:
agent-budget serve
Or use it programmatically:
from agent_budget.mcp_server import mcp
mcp.run()
MCP Tools
Budget Tools
create_budget— Create a new budgetlist_budgets— List all budgetsget_budget— Get budget detailsupdate_budget— Update a budget's settingsdelete_budget— Delete a budgetprocess_budget_rollover— Carry unspent budget forwardget_budget_status— Check actual vs. budgeted spendingcompare_budget_actual— Detailed budget comparisonupdate_alert_thresholds— Customize alert thresholds
Expense Tools
add_expense— Log a new expenseupdate_expense— Update an existing expenselist_expenses— List expenses with filters (category, vendor, reimbursable, etc.)get_expense— Get expense detailsdelete_expense— Delete an expense
Savings Goal Tools
create_savings_goal— Create a savings targetlist_savings_goals— List savings goalsget_savings_goal— Get goal details with progresscontribute_to_savings— Add a contributionwithdraw_from_savings— Withdraw from a goalupdate_savings_goal— Update a goaldelete_savings_goal— Delete a goal
Spending Rule Tools
create_spending_rule— Create a spending control rulelist_spending_rules— List spending rulescheck_expense_rules— Check if an expense would violate rulesupdate_spending_rule— Update a ruledelete_spending_rule— Delete a rule
Recurring Expense Tools
add_recurring_expense— Set up a recurring expenselist_recurring_expenses— List recurring templatesprocess_recurring_expenses— Generate expenses from due templates
Analysis Tools
get_spending_forecast— Project future spendingget_spending_summary— Spending by categoryget_alerts— Check budget alertsclear_alerts— Clear alertsexport_data— Export all datalist_currencies— List supported currencies
Python API
from agent_budget.service import BudgetService
from agent_budget.store import BudgetStore
from agent_budget.models import BudgetPeriod, RecurringFrequency, SpendingRuleAction
# Initialize
svc = BudgetService(BudgetStore())
# Create a budget with rollover
budget = svc.create_budget(
name="API Costs",
limit=500,
period=BudgetPeriod.MONTHLY,
category="api",
rollover_enabled=True,
rollover_cap=100,
)
# Add an expense
expense = svc.add_expense(
amount=25.50,
category="api",
description="OpenAI GPT-4 call",
vendor="OpenAI",
budget_id=budget.id,
)
# Update an expense
svc.update_expense(expense.id, amount=30.00, receipt_url="https://receipt.example.com/123")
# Create a savings goal
goal = svc.create_savings_goal(
name="Emergency Fund",
target_amount=10000,
target_date=date(2027, 1, 1),
)
# Contribute to the goal
goal = svc.contribute_to_savings(goal.id, amount=500, note="Monthly deposit")
# Create a spending rule
rule = svc.create_spending_rule(
name="API Cap",
category="api",
action=SpendingRuleAction.BLOCK,
threshold_amount=500,
)
# Check budget status
status = svc.get_budget_status(budget.id)
print(f"Used {status.percent_used}% of budget")
# Process budget rollovers
results = svc.process_all_rollovers()
# Get spending forecast
forecasts = svc.get_spending_forecast(months=3)
Data Storage
All data is stored in JSON files under ~/.agent-budget/ (or the directory specified by the AGENT_BUDGET_DIR environment variable). No external database required.
Supported Currencies
USD, EUR, GBP, JPY, CAD, AUD, CHF, CNY, INR, BRL, KRW, MXN, SGD, SEK, NZD
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。