ProofDiff · evidence report

What evidence exists for this change?

A local, evidence-based view of changed code, related tests, deterministic checks, and the places that still need human judgment.

acme-checkoutworking tree vs HEAD2026-08-12T00:00:00.000Zb39d3dbbd762

Partially verified

2 changed files · 2 changed symbols · highest risk high

Evidence strength differs across changed files. Inspect each file before drawing a conclusion.

Change review queue

Highest-risk files appear first. Expand a file to audit each claim.

services/email.py

modified · +7 −2 · python

Unverifiedhigh risk · 58
function · send_receipt

Checks ran, but supplied no applicable successful evidence for this file.

Inspect evidence and reasoning

Evidence

  • No applicable check

    Checks ran, but none could be associated with this file's language.

    high confidence · limitation
  • 2 call references in changed lines

    Parser-observed name-only call sites: range, client.send. Targets are not resolved and runtime execution is not implied.

    high confidence · static relationship
  • python3 ast structural analysis

    Parser confidence: high. Structural findings are evidence about code shape, not correctness.

    high confidence · inference
  • Evidence boundary · static-relationship · no-related-test

    ProofDiff established no supported related test-like path for this change. Next action: Inspect static-analysis limitations and test relationships; absence of a discovered relationship is not proof that no relevant test exists.

    high confidence · limitation

Why review this

  • Executed checks provided no applicable successful evidence.
  • No statically related test-like path was found.

Targeted test outcomes

None observed. Related-file presence alone cannot produce “Related test file passed.”

Statically related test-like paths

None found through resolved static imports or runner qualification.

Calls in changed lines 2

  • range line 2 · high confidence
  • client.send line 4 · high confidence

Name-only structural references; targets are not resolved and runtime execution is not implied.

Estimated impact

No dependent files resolved.

Limits

  • Deleted symbol names were inferred from removed declaration lines; ranges and full structure are unavailable.
  • No test-like path or runner-qualified target was related to the change; dynamic imports and runtime dispatch may not be visible statically.

src/discount.js

modified · +3 −2 · javascript

Related test file passedlow risk · 10
function · discountFor

Observed: a statically related, runner-qualified target was explicitly supplied and produced at least one non-skipped passing test for that exact path. Not observed: changed-symbol or changed-line execution.

Inspect evidence and reasoning

Evidence

  • test: test

    Repository test command passed, but ProofDiff did not observe which test files it executed.

    high confidence · passing check
  • 1 qualified related target observed passing

    ProofDiff explicitly supplied each qualified target and observed at least one non-skipped passing test for that exact path. This is file-scoped test evidence, not changed-symbol, changed-line, branch, assertion, or behavioral coverage.

    high confidence · executed test
  • 2 call references in changed lines

    Parser-observed name-only call sites: Math.max, Number. Targets are not resolved and runtime execution is not implied.

    high confidence · static relationship
  • 2 impacted files

    Reachable through resolved reverse import relationships. This is an impact estimate, not proof of runtime behavior.

    medium confidence · static relationship
  • @babel/parser structural analysis

    Parser confidence: high. Structural findings are evidence about code shape, not correctness.

    high confidence · inference
  • Evidence boundary · changed-code-execution · changed-code-execution-unobserved

    A runner-qualified related test file passed, but ProofDiff did not observe whether changed symbols, lines, branches, or relevant assertions executed. ProofDiff intentionally failed closed at this boundary.

    high confidence · limitation

Why review this

  • No additional risk factors identified.

Targeted test outcomes 1

  • passed test/checkout.test.js

Statically related test-like paths 1

  • test/checkout.test.js

Calls in changed lines 2

  • Math.max line 2 · high confidence
  • Number line 2 · high confidence

Name-only structural references; targets are not resolved and runtime execution is not implied.

Estimated impact 2

  • src/checkout.js
  • test/checkout.test.js

Limits

  • Qualified related targets produced passing tests, but ProofDiff did not observe whether changed symbols, lines, branches, or relevant assertions executed.

Verification checks

Commands, outcomes, and bounded output from this run.

passedtest: test146 ms

The command exited successfully. package.json script "test".

npm run test --silent
✔ gold customers receive the documented discount (0.303292ms)
✔ unknown tiers pay the full subtotal (0.050083ms)
ℹ tests 2
ℹ suites 0
ℹ pass 2
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 40.689875
passedtargeted node-test: 1 qualified test target72 ms

The command exited successfully. Target observations: 1 passed, 0 failed, 0 skipped, 0 zero-test, 0 unavailable. ProofDiff targeted execution derived from package.json script "test".

node --test --test-reporter=<ProofDiff observer> test/checkout.test.js

Target qualifications

  • test/checkout.test.jstest/checkout.test.js · runner-default-pattern (high)
    test/checkout.test.js matches a documented default node --test discovery pattern. Default-pattern qualification establishes runner identity, not test execution or behavioral coverage.

Target observations

  • passed test/checkout.test.js · 2 non-skipped observed
    The runner observed 2 passed, 0 failed, and 0 skipped tests for this exact target.
✔ gold customers receive the documented discount (0.307375ms)
✔ unknown tiers pay the full subtotal (0.050292ms)
ℹ tests 2
ℹ suites 0
ℹ pass 2
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 39.004667

How to read this: The JSON status verified is displayed as “Related test file passed.” It requires a static relationship, runner qualification, explicit target supply, and at least one non-skipped passing test observed for that exact path—not changed-symbol or changed-line execution, mathematical proof, or a guarantee of safety.