Global Bond Agent

Agentic fixed-income screening across sovereigns, credit, munis and EM · signals blend technical + macro-narrative · all P&L net of frictions
Individual bonds have no ticker, only a CUSIP. Type a bond ETF ticker and we will show you the market it tracks.
Start here — the short version
How much of the world’s bond market does this cover?

Every bond that exists is not a realistic target, and it is worth being clear why. There are millions of individual bonds outstanding. The 500-odd non-financial companies in the S&P 500 alone have more than 7,000 separate bonds between them, and one large bank can have tens of thousands. Add every government, every agency, and roughly a million US municipal bonds, and no dashboard — free or paid — lists them all one by one. The databases that come closest cost tens of thousands a year, and even they cannot price most individual bonds, because most bonds do not trade on any given day.

What this does instead is cover the market by value rather than by count. The 102 entries here span every major government bond market, every credit rating from AAA down to CCC at every common length, municipals, mortgage and asset-backed debt, and emerging markets in both dollars and local currency. Between them they account for the large majority of the money actually invested in bonds worldwide. Any individual bond you might buy sits inside one of these buckets and behaves like it.

For one specific bond, paste its CUSIP or ISIN into the search box at the top. It points you to the free public registries that genuinely are complete: FINRA carries every US corporate and agency bond trade, and EMMA carries every US municipal bond with prices and disclosures. Both free. That is the honest route to bond-by-bond coverage: this dashboard tells you which part of the market to be in, those registries tell you what a particular bond in it is worth.

Every investment, ranked
InstrumentPays youExtra vs govt LengthRate sensitivityScore ActionModel expects Before costsYou keepYou keep, $ Reward vs swingsConfidence
Green — the model expects the price to rise Red — the model expects the price to fall Net = what you actually keep, after Cyprus tax and the dealer’s cut. Tap or click any row for the full breakdown

Why the best-rated ones screen well

Yield history — the model reads momentum, trend and valuation range off this series.

Why the model rates it this way

What has happened to it so far

Expected P&L — 12 month hold

If you were to act on this

What if rates move?

Net return if yields move by the amount shown, over the selected horizon.

Where this number comes from

What could go wrong

US Treasury curve

Live from Treasury.gov daily par yield curve. Today vs 1 month and 6 months ago.

Curve slopes

2s10s and 10s30s. Steepening = term premium rebuilding; long end underperforms.

Global 10-year sovereign yields

Credit spreads vs history

Current OAS against 10-year percentile band. Low percentile = expensive credit.

Macro narrative inputs

Build a portfolio and watch how it does

Add what you own, or what you are thinking of buying. This page then values it every day using real market history and shows what it would have done. Everything is saved on this device only — nothing is sent anywhere, and no account is needed.

What it has been worth

Green is what your holdings are worth including interest collected. Grey is what you put in. The gap between them is your profit or loss.
Valued using the actual daily market history behind this dashboard, so the path is real rather than a straight line. Two simplifications worth knowing: interest is counted as it accrues rather than on the exact payment dates, and the dealer’s cut is not deducted until you actually sell.

Let the model build a portfolio for you

Equal-risk weighting: each position is sized so its duration contribution is equal, then scaled to capital. This keeps a 30-year bond from dominating the book's rate risk.

Book allocation

Return attribution

Ask about these bonds
Not investment advice. This is a research and screening tool. Signal scores and expected returns are model output, not forecasts — they describe what the stated rules imply given current data, and the rules can be wrong. Bond-level pricing here is benchmark/index level, not executable quotes; real fills carry bid-offer that can exceed the modelled friction, especially in munis, high yield and EM. Verify with your broker before trading.