Kalkasautonomous decisions
All domain modules

Knowledge-dated, restatement-aware

Financial analysis

Analytical stances that respect knowledge dates, restatements, and survivorship.

Run it locallyModule referenceWalk the loop

Evidence

What this module supplies.

Facts are admitted at the date the market could have known them, not the date the report was printed. Restatements and delistings are modelled rather than papered over, and the run ends in an analysis stance with a report projection — never an order.

Gates

What gates every run.

Gates are evaluated before a decision is emitted, not at execution. A run that cannot clear one ends in a reasoned abstention or a rejected candidate instead of an effect.

Limits

What this module never does.

These are contract lines rather than caveats: the module has no path that requests them, in any configuration, on any surface.

  • neverNo order path and no execution adapter
  • neverNo recommendation presented as advice
  • neverNo figure admitted before it was knowable

LimitProduct-wide, in every module

  • · No live decision, order, wager, or external effect exists in the shipped adapter.
  • · No custody of funds, accounts, or customer assets, and no counterparty reach-through.
  • · No guaranteed profit, no promised strike rate, and no performance claim sold as fact.
  • · No action when evidence, authority, replay, settlement, or effect identity is incomplete.

Worked sample run

The record this module produces.

An illustrative sample from this module, with the charts reading the numbers the run reports. The values are samples, and no production run exists yet.

sample dataRestated margin trajectory — listed industrial
decision clock 22 Jan 2026, 09:15 UTCReasoned abstention

Restatement moved the knowledge date past the decision clock, so the run abstained with a reason — and settled as a correct abstention.

belief

0.00belief density1.00
p = 0.54interval 0.38–0.70interval width 32%

Analytical interval model v2.1.0 — Restatement widened the interval past the confidence floor for an analysis stance.

believed against observed, sealed holdout

0.50.60.70.80.9
believed (x) against observed frequency (y) · dashed line is perfect calibration

evidence admitted

  • Reported figuresFilings index · knowledge-dated
    observed 20 Jan 2026, 07:00 UTC · available 20 Jan 2026, 07:00 UTCMargins admitted at the date the market could have known them.
  • Restatement noticeFilings index · amendment
    observed 22 Jan 2026, 08:10 UTC · available 22 Jan 2026, 08:12 UTCAmendment landed after the belief was formed; inference was re-run, not patched.
  • Peer multiple setSurvivorship-aware peer set
    observed 21 Jan 2026, 18:00 UTC · available 21 Jan 2026, 18:05 UTCDelisted peers retained so the comparison set does not flatter the result.

mandate checks

Objective: Analysis-only stances for the listed-equity universe

  • Mandate presentSigned, unexpired authoritypass
  • Confidence floorInterval must exclude 0.50refuse
  • Restatement awarenessAmendment applied before stancepass
  • Permitted effectAnalysis stance only — no order pathpass

authority expires 31 Mar 2026, 00:00 UTC

decision

DecisionReasoned abstention — confidence too low

After the restatement the interval straddled the decision boundary. The run abstained rather than shipping a stance the evidence did not support.

effect and settlement

No adapter was reached for this run: nothing was planned, so there is nothing to retry and nothing to undo.

settlement
Outcome recorded against the abstention; no stance was published.
evaluation
Abstention counted in coverage rather than dropped.
learning
Learning receipt: restatement lag, not the model, was the binding constraint.

run fingerprint sha256:2ce7b104…90afreplay deterministiclive:false

Illustrative worked example. Field names match a real run receipt; the values are samples, and no production run exists yet.

Next steps

Keep going from here.

Read the contract, run the loop, then read what the run recorded.