Most “AI literacy” products are slideshows with a quiz at the end.
We already run a live agent economy — Hubs that index capabilities, peers that join and leave, prices that move, provenance that either exists or doesn’t. Operators keep misreading that glass: a green pin that means SIM, a quiet peer that means gone, a catalogue that looks rich because one source is dumping tools.
So we wrapped the real telemetry in a hunt.
Signal Hunt is a federation-native detective game — and an educational lab. Every round starts from a measured Hub snapshot. You open evidence (it costs score). You commit a diagnosis and a confidence. The server reveals a precommitted salt, recomputes a Brier score, and signs the milestone.
No fixtures. No placeholder catalogue. If the Hub cannot be observed, the round is unavailable — not “all clear.”
Play: hunt.modelmarket.dev · Source: github.com/alexar76/signal-hunt (MIT) · Landing: alexar76.github.io/signal-hunt
The thesis (one sentence)
Federation literacy is a skill — and skills get worse when the tutorial invents the world.
Green UI is cosplay unless the numbers come from a living Hub. Fake confidence is worse than a wrong call. Signal Hunt punishes both.

Landing: LIVE TELEMETRY · Ed25519 · Brier · measured state only.
Four moves. One verifiable verdict.
The protocol is deliberately short:
| Move | What happens |
|---|---|
| 01 Observe | Node records a live Hub manifest — sources, prices, provenance, state hash |
| 02 Investigate | Open only the evidence you need; each reveal is committed and costs score |
| 03 Commit | One diagnosis + confidence (25–100%). One submission per round |
| 04 Verify | Salt revealed. Brier recomputed. Milestone signed |

Nothing in the result depends on a hidden moderator or invented telemetry.
Six evidence blocks sit behind the glass: distribution · historical change · pricing · peers · latency · provenance. Curiosity is not free. That is the pedagogy.
A sealed field reading is ready
The detector class stays hidden until you commit. The answer was salt-committed before you acted. You are not negotiating with a GM — you are racing a precommitted world.

Round sealed. Open evidence. Choose a diagnosis. Verify against the commitment.
Diagnoses map to real federation dynamics — isolation, disappearance, peer churn, latency weather, concentration, and friends. After a few rounds you should be able to say why a Hub looks the way it does from measured sources, not from vibes.
Fake confidence loses on purpose
Correctness alone is not enough. You state a probability. The server scores you with Brier. Calibrated confidence is rewarded. Bluffing certainty is punished. Math — not a moderator.

Salt revealed. Commitment independently verifiable. Score is earned, not narrated.
When a verdict lands, you can recompute it from salt, commitment, and returned operands. That is the whole point of sealing the answer early: reproduce the result.
Astral progression — cosmetics, not money
Progression is astral: Stargazer toward Pathfinder. Score comes from verified rounds. A season passport tracks the week — distinct diagnoses, PRIME corrects, streaks.
Relics are deterministic records. Nothing is minted. Cosmetics only. Empty hero feed is honest: nobody opted in yet — not a fake leaderboard.

Climb by being right — and calibrated. Season passport is a ledger of skill, not a wallet.
Heroes of the signal only publish opt-in milestones from signed, verified verdicts. DIOSCURI can relay to Discord / X — only with explicit consent. Share proof. Never invent impact.
Why this sits next to the rest of the economy
Signal Hunt is not a side quest bolted onto a blog.
It runs against an ordinary AIMarket Hub (same family as the competing lab / USE portal). The game registers local capabilities (signal.case, signal.evidence, signal.submit, …) and discovers remote tools through the Hub when they exist. Failure is recorded as unavailable — never dressed up as a successful call.
If you already know ATLAS (LIVE needs provenance; SIM stays labeled) and GAIA (silence → $0), this is the same honesty grammar — aimed at operators reading federation glass, not at settlement physics.
Crypto stays optional for the lesson. The shape is the product: observe → pay for evidence → commit → verify.
Soft star
If this is your kind of lab — a detective loop on live rails instead of a fixture tutorial — a star helps the repo travel.
⭐ github.com/alexar76/signal-hunt · Play: hunt.modelmarket.dev
The federation emitted a signal.
Find the cause.

Top comments (0)