monthly-money-review
Skill0 starsChanged 18 days agoMITView on GitHub
Orchestrate a complete monthly financial review using BankBridge: cashflow, top categories, merchants, subscriptions, investment delta, narrative takeaways. Use when the user asks for a monthly check-in, a "how am I doing", or a broad look at their money for a specific month.
In this repository5 other files it gives its agents.
- budget-builderSkill
- fraud-detectiveSkill
- investment-health-checkSkill
- subscription-auditorSkill
- tax-prep-assistantSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Use this skill when the user says something like:
- Distinctive: line 20
Pick the month: If the user named one ("March", "2026-03", "last month"), use that.
- Distinctive: line 22
Gather the data: Call in this order:
- Distinctive: line 63
Write in the user's voice: second person ("you spent..."), not third ("the user spent...").
- Distinctive: line 65
Keep takeaways honest: if things are fine, say so briefly.
- Very specific: line 66
If BankBridge returns a warning (e.g.
- Distinctive: line 70
Do not suggest specific subscription cancellations unless the user asks.
- Distinctive: line 71
Do not pattern-match emotionally charged categories (medical, debt collections, gambling).
- Distinctive: line 72
Do not compare against "averages".
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Monthly money review
You have access to BankBridge MCP tools. The user wants a complete picture of their finances for a specific month.
When to use
Use this skill when the user says something like:
- "how am I doing financially this month"
- "give me a monthly review / check-in / rundown"
- "summarize my money for March"
- "where did my money go last month"
Process
- Pick the month. If the user named one ("March", "2026-03", "last month"), use that. Otherwise default to the most recent complete calendar month. Never use the current in-progress month unless they ask specifically; partial data skews the narrative.
- Gather the data. Call in this order:
get_monthly_cashflowfor the target month → income, expenses, net, top sources, top categoriesget_spending_summarygrouped bycategoryfor the same date range → full category breakdownget_spending_summarygrouped bymerchant, limit 10 → top merchantsget_recurring_chargeswith date range covering the target month → list of recurring charges active that monthlist_holdings+list_investment_transactionsfiltered to the target month if the user has a brokerage account connected (skip otherwise)
- Compose the report in this exact structure. One page of well-formatted markdown:
# Monthly money review: <Month Year>
## Cashflow
- Income: $X,XXX
- Expenses: $X,XXX
- Net: $X,XXX (label: "saved" if positive, "over-spent" if negative)
- Savings rate: X% of income (only if income > 0)
## Top 5 spending categories
| Category | Spend | % of total |
|---|---|---|
## Top 10 merchants
| Merchant | Spend | Visits |
|---|---|---|
## Recurring charges
List every recurring charge with amount, frequency, total monthly impact. Call out the monthly subtotal.
## Investment delta (if any brokerage connected)
- Total portfolio value: $X,XXX
- Gain/loss this month: $X,XXX
- Biggest mover: TICKER +/-X%
- Dividends received: $XXX
## Takeaways
Two to four short, concrete observations. Prioritize things the user might not already know: surprise expenses, category shifts vs. the month before (use get_monthly_cashflow on the prior month if useful for comparison), subscriptions that crept up, a merchant they visited unusually often, anything genuinely actionable.Style
- Write in the user's voice: second person ("you spent..."), not third ("the user spent...").
- Real numbers from the tools only. Never estimate or fabricate.
- Keep takeaways honest: if things are fine, say so briefly. No fake urgency.
- If BankBridge returns a warning (e.g. REAUTH_REQUIRED, NO_CONNECTIONS), relay it to the user before proceeding or in place of the report.
Guardrails
- Do not suggest specific subscription cancellations unless the user asks. This skill is a report, not a recommendation engine.
- Do not pattern-match emotionally charged categories (medical, debt collections, gambling). Just report numbers.
- Do not compare against "averages"; you don't have population data. Compare against the user's own prior months only.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

