agentleFS
bankbridge-moneybankbridge-money/bankbridge-skills/investment-health-check/SKILL.md

investment-health-check

Skill0 starsChanged 18 days agoMITView on GitHub

Portfolio snapshot across connected brokerage accounts: total value, gain/loss by position, concentration warnings, dividend run-rate. Use when the user asks about their investments, portfolio, holdings, gains, or dividends.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Confirm there's an investment account connected: Call list_accounts. If none has type: "investment", tell the user:

    line 22
  2. Distinctive:

    Don't offer financial advice ("you should rebalance") unless asked.

    line 73
  3. Very specific:

    If the user asks about a specific ticker, use get_merchant_history with the ticker name, though note that investment transactions and …

    line 74
  4. Distinctive:

    Do NOT make tax claims ("this is a long-term capital gain").

    line 78
  5. Distinctive:

    Do NOT predict returns.

    line 79

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Investment health check

You have access to BankBridge MCP tools. The user wants a clear view of their investment positions and how they're performing.

When to use

Trigger on:

  • "how's my portfolio"
  • "what are my holdings"
  • "am I up or down this year"
  • "what are my biggest positions"
  • "how much have I received in dividends"
  • "am I too concentrated in any stock"

Process

  1. Confirm there's an investment account connected. Call list_accounts. If none has type: "investment", tell the user:

    "I don't see a brokerage connected. If you have one with Fidelity / Schwab / Vanguard / Robinhood / etc., add it via connect_bank or the dashboard and I'll run this again."

    Stop here.

  2. Pull holdings + recent activity:
    • list_holdings → current positions with gain/loss
    • list_investment_transactions with start_date 365 days ago → buys/sells/dividends for the year
    • Optionally filter subtype: "dividend" for a clean dividends cut
  3. Compute key numbers:
    • Total value = sum of value across holdings
    • Total gain/loss (dollar) = sum of gain_loss for positions with cost_basis != null. Flag any positions without cost basis, common for long-held positions the brokerage doesn't report.
    • Total gain/loss (%) = gain/loss $ / total cost basis $, when denominator is non-zero
    • Concentration = each position's value / total_value. Flag anything > 20% as concentration risk.
    • YTD dividends = sum of absolute amount for all transactions where subtype == "dividend" and date is in current calendar year
    • Run rate = (YTD dividends / days elapsed YTD) × 365
  4. Compose the report:
# Investment health check: <today>

## Portfolio at a glance
- Total value: $XXX,XXX
- Total gain/loss: $XX,XXX (+X.XX%)  ← color or "+" for gain, "-" for loss
- Cost basis: $XX,XXX (coverage: X of N positions)

## Positions
Table: ticker | shares | value | gain/loss $ | gain/loss % | % of portfolio
Sort by value descending.

## Concentration
If any position > 20% of portfolio:
  "<ticker> is XX% of your portfolio. Concentration risk: consider whether that's intentional."
Otherwise:
  "Biggest position: <ticker> at XX%. No single stock over 20%."

## Dividends
- YTD received: $X,XXX
- Run-rate projection: $X,XXX / year
- Top 3 payers: <ticker $X, ticker $X, ticker $X>

## Activity last 90 days
Any buys / sells / fees: brief summary.

Style

  • Lead with the number (value), then the trend (gain/loss), then the concerns (concentration).
  • Don't offer financial advice ("you should rebalance") unless asked.
  • If the user asks about a specific ticker, use get_merchant_history with the ticker name, though note that investment transactions and merchant history are separate surfaces. Stick to list_investment_transactions filtered by security_id when possible.

Guardrails

  • Do NOT make tax claims ("this is a long-term capital gain"); you don't know acquisition dates at lot level.
  • Do NOT predict returns.
  • Flag positions with cost_basis = null clearly. Don't include them in percentage-based totals.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.