ShaAnas

ShaAnas / SignalProof 2.0

Same event count.
Different data.

Compare supplied JSON input and output against explicit filter and mask rules. Record identities, copy counts, types and values can reveal a mismatch that a total count misses.

Private evaluation · implementation access by agreement
Request a private pilotView illustrative cases

This public page presents modeled examples. It does not accept customer files, run a live check or connect to your infrastructure.

ILLUSTRATION / NOT A LIVE RUN
Expected: 80 recordsA modeled filter-and-mask result
Observed: 80 recordsOne missing record and one extra copy can balance
Equal totals are insufficientIdentity, multiplicity, types and values matter

The examples below describe questions for an agreed evaluation, not newly measured results.

Three modeled cases

See the problem.
Agree what to verify.

These are illustrative scenarios, not live results or claims that this page has tested a real configuration or pipeline.

01 / ILLUSTRATION

The expected records match.

A modeled output contains the identities, copy counts, JSON types and values required by the declared rules.

Question for a private evaluation: Does the supplied output match the expected record set, even if record order changes?

Modeled example · no live check
02 / ILLUSTRATION

A loss and a duplicate balance.

The modeled expected and observed totals are both 80, but one record is missing and another appears an extra time.

Question for a private evaluation: Which identities and copy counts differ despite the equal totals?

Modeled example · no live check
03 / ILLUSTRATION

The masking rule was not reflected.

A rule expects a top-level field to be replaced with a fixed mask, while the modeled output retains a different value.

Question for a private evaluation: Does the actual field value agree with the declared transformation?

Modeled example · no live check

Private pilot scope

A focused check,
with agreed evidence.

A focused evaluation can compare an agreed pair of JSON input/output files against supported filter-to-mask rules, reproduce the relevant example and deliver a report. Data handling and expected checks are agreed before files are shared.

Agree the materials, environment, expected outcomes and access before evaluation. The implementation and evaluation components are supplied privately by agreement.

Read the limits with the result.

  • The supported model uses a top-level equality filter followed by fixed-value masking; aggregation and arbitrary transformation code are outside that model.
  • Record order is not verified by the record-set comparison.
  • Supplied JSON and fingerprints do not prove external delivery, data provenance or the behaviour of an unconnected production pipeline.

Discuss a small, concrete case

Start with the task.

Describe the problem and the result you need to verify. Please agree data handling with us before sharing sensitive files or credentials.

Request a SignalProof 2.0 pilotView other products