Skip to main content
Kam AIDocs
These help docs are for Kam AI users, future customers, and AI agents learning what the product does. For a machine-readable summary, see /llms.txt.
Kam AIHelp center

How can we help?

Learn how to use Kam AI for source aggregation, signal review, freshness checks, saved ideas, and better follow-up questions.

Popular topics

Getting startedPromptsBoard scanSource reviewSignal reviewFreshnessMarket journalTrustOutcome reviewTroubleshooting

The manifesto

Sports-market research should make non-technical analysts more powerful.

Kam AI exists because the future of sports research is not another static dashboard. It is a sharper workflow for aggregating fragmented sources, asking plain-English questions, saving ideas, and learning from outcomes.

Get early accessExplore playbooks

Research system

3-part prompt

Object, signal, question.

5-part answer

Source, change, why, caveat, next check.

4 freshness states

Fresh, stale, delayed, unavailable.

The opportunity

Sports-market research needs memory, not louder dashboards.

AI can make research faster, but speed only helps when the process stays grounded. Kam is built to help users move from raw source fragments to a saved thesis, source check, and outcome review in one place.

Sources move fast

Kam helps you ask what changed, why it matters, and whether the context is still fresh enough to use.

Most answers lack memory

Kam connects the source, signal, saved reason, and next check so you can judge the reasoning.

Research gets scattered

Use one agent workflow for odds, news, injuries, schedules, stats, watchlists, and outcome review.

Core workflows

One chat flow for the full research cycle.

Ask better questions

Give Kam the source, watched object, signal, or saved idea you want to understand.

Understand the signal

Use Kam to explain what changed, which source changed, and whether context is fresh.

Read with discipline

Look for source context, caveats, missing data, and what would change the read.

Verify freshness

Check whether data is fresh, stale, delayed, or unavailable before trusting an answer.

How Kam works

Use the chat screen to give Kam a research job. Kam works best when you include the sport, watched object, signal, source, or saved idea you are trying to understand.

Kam is not a transaction engine or outcome guarantee. It helps you reason through sports-market information, but you still verify live inputs and make the final decision.

Ask

State the object, signal, source, or saved idea.

Read

Look for source, change, caveats, and missing data.

Verify

Check freshness and late context before deciding.

What to ask

Kam is useful when the question has a real information problem behind it. Ask it to test a thesis, compare source context, find risk, or explain what changed.

JobUse Kam to
Board scanFind watched markets with meaningful changes, stale sources, delayed feeds, or unresolved signals.
Signal explanationAsk what changed, what the prior value was, why it matters, and what needs verification.
Freshness checkConfirm whether data is fresh, stale, delayed, or unavailable before using the answer.
Save market ideaCapture the reason, source snapshot, expected movement, and what would change the view.
Outcome reviewReview whether the reason held up, whether the market moved as expected, and what lesson to save.
Portfolio reviewLook for repeated assumptions, stale reasons, unresolved ideas, and correlated context.

Trust standard

Kam users are analytical, often non-technical, and need sober judgment under changing conditions. Trust comes from clear reasoning, visible caveats, freshness labels, and grounded language.

  • No guaranteed outcomesKam is for research, education, and decision support. It does not promise results.
  • Evidence firstA useful answer should show the main reasons behind the read.
  • Caveats matterMissing injuries, stale sources, small samples, and source conflicts should be called out.
  • Freshness mattersSports markets move fast. Kam should mark fresh, stale, delayed, and unavailable inputs.
  • You make the callKam can sharpen a thesis, but the final decision is yours.

Good prompts

A good Kam prompt names the object, signal or source, and the exact question. Short is fine if the key facts are there.

Board scan“Scan today’s NBA board. Which watched markets changed materially, and which signals are stale or delayed?”
Signal“Explain this line_move signal. What source changed, what was the previous value, and what should I verify?”
Save idea“Save this market idea with my reason: I am watching Knicks spread because injury timing may be overreacted to. What needs to be true by close?”
Review“Review this closed idea. Did the reason hold up, did market movement support it, and what lesson should I save?”
Plain English“I am not technical. Explain the source context behind this signal like a stats analyst, not a developer.”

Playbooks

Use these patterns when you are not sure how to frame the question. They keep Kam focused on what changes the research conclusion.

Board scan

Ask which watched objects changed, which sources are stale, and which signals deserve attention first.

Source review

Ask where a data point came from, when it last changed, and whether there are source conflicts.

Signal review

Ask what changed, what the prior state was, and why the signal matters to the saved idea.

Market journal

Ask Kam to save the reason, source snapshot, next check, and what would change your view.

Outcome review

Ask Kam to separate good process from stale data, bad assumptions, and normal uncertainty.

Reading answers

Read Kam answers like a research note. The most important line is often the caveat or freshness state, not the summary.

SummaryThe short read. Treat it as a research note, not a command.
SourceWhere the answer is grounded and which data changed.
CaveatWhat can break the thesis. This is often the most valuable part.
Missing dataWhat Kam does not know yet or what may be stale.
Next checkThe one thing to verify before making a decision.

Verification checklist

Before trusting any research idea, slow down and verify the inputs that change fastest.

  • Check whether the source is fresh, stale, delayed, or unavailable.
  • Recheck injuries, starters, weather, scratches, schedule changes, and late news.
  • Ask what missing data would change the answer.
  • Look for repeated assumptions across saved ideas.
  • Do not treat last-known data as current truth.
  • Keep research confidence separate from guaranteed outcomes.

Common mistakes

Most weak results come from vague prompts, stale inputs, or treating a summary like a guarantee.

Asking for an answer onlyAsk for the source, signal, caveat, and next check instead.
Ignoring stale dataA good read from old context can become a weak read when inputs change.
Overvaluing recent gamesAsk Kam to balance recent form with role, matchup, and sample size.
Skipping the counterargumentAsk what would make the saved idea weak or underspecified.
Stacking assumptionsSeveral ideas can depend on the same hidden source or game script. Ask Kam to check that.

Troubleshooting

If the answer is not useful, tighten the question. Kam performs better when the decision is clear.

The answer is too broadAdd league, watched object, signal, source, and what you are trying to understand.
The answer feels too confidentAsk Kam for caveats, missing data, freshness state, and the strongest counterargument.
The data may be staleAsk what needs a fresh check, then verify current source context outside the answer.
You disagree with KamAsk it to argue the other side and list what evidence would change its read.
You are short on timeAsk for a 5-bullet brief with source, change, caveat, missing data, and next check.

Ready to research smarter?

Start with one clear object, signal, source, or saved idea.

Get early access