Verify Claim
Verification proves or disproves one claim; it is not a progress recap.
Workflow
- Restate the claim as condition, measured outcome, threshold, and evidence tier.
- Select a valid baseline/treatment or condition/falsifying-control comparison. Freeze
source or artifact identity, input population, checker, environment, public entrypoint,
process/window/session identity, warmup, units, timeout, and visible baseline.
- Run the cheapest check that can falsify the claim. Preserve raw evidence or immutable
identity before summarizing it.
- Use the same checker in both arms. Identical arms, universal zeroes, missing inputs,
setup errors, and unexpected skips are instrumentation alarms.
- Preserve intermediate states; a correct final frame cannot erase an earlier visible
or same-event failure.
- Return exactly
VERIFIED, NOT VERIFIED, or INCONCLUSIVE.
Evidence boundaries
Keep source tests, package-shaped tests, native/ABI, immutable artifacts, real UI, author
operation, upload/readback, and production evidence distinct. A hidden launcher, mock,
synthetic widget, or owner-written screenshot cannot replace a claim about a public entry,
real process, or user-visible output. A skip, timeout, setup error, or missing baseline is
not a pass.
Preserve the user's reported event order and sample provenance. A missing event in an
incomplete export does not prove it never occurred; a later review screen does not prove
an earlier completion screen was absent. Align session, version, path, and timestamps.
Keep canonical local evidence and reproducibility metadata within the selected workflow.
Paths, hashes, filenames, and non-secret identifiers are not secret by shape alone; derive
external views by content and destination without rewriting the original evidence.
Use bounded reads; failed or partial lookup cannot prove absence. A verified finding
does not by itself authorize a fix or a new release requirement.
Output
Return the claim, verdict, comparison identities, matched metrics/artifacts, threshold or
delta, confounds, and not-run. If only static inspection is authorized, label it clearly
and keep the verdict INCONCLUSIVE unless the claim itself is static.
For a matched installed-CLI example and verdict boundaries, read
the synthetic example.
1---2name: verify-claim3description: Verify one falsifiable behavior or performance claim with matched baseline and treatment evidence. Use for before/after fixes, package behavior, UI, startup, or workflow claims; not for broad audits.4---56# Verify Claim78Verification proves or disproves one claim; it is not a progress recap.910## Workflow11121. Restate the claim as condition, measured outcome, threshold, and evidence tier.132. Select a valid baseline/treatment or condition/falsifying-control comparison. Freeze14 source or artifact identity, input population, checker, environment, public entrypoint,15 process/window/session identity, warmup, units, timeout, and visible baseline.163. Run the cheapest check that can falsify the claim. Preserve raw evidence or immutable17 identity before summarizing it.184. Use the same checker in both arms. Identical arms, universal zeroes, missing inputs,19 setup errors, and unexpected skips are instrumentation alarms.205. Preserve intermediate states; a correct final frame cannot erase an earlier visible21 or same-event failure.226. Return exactly `VERIFIED`, `NOT VERIFIED`, or `INCONCLUSIVE`.2324## Evidence boundaries2526Keep source tests, package-shaped tests, native/ABI, immutable artifacts, real UI, author27operation, upload/readback, and production evidence distinct. A hidden launcher, mock,28synthetic widget, or owner-written screenshot cannot replace a claim about a public entry,29real process, or user-visible output. A skip, timeout, setup error, or missing baseline is30not a pass.3132Preserve the user's reported event order and sample provenance. A missing event in an33incomplete export does not prove it never occurred; a later review screen does not prove34an earlier completion screen was absent. Align session, version, path, and timestamps.3536Keep canonical local evidence and reproducibility metadata within the selected workflow.37Paths, hashes, filenames, and non-secret identifiers are not secret by shape alone; derive38external views by content and destination without rewriting the original evidence.39Use bounded reads; failed or partial lookup cannot prove absence. A verified finding40does not by itself authorize a fix or a new release requirement.4142## Output4344Return the claim, verdict, comparison identities, matched metrics/artifacts, threshold or45delta, confounds, and `not-run`. If only static inspection is authorized, label it clearly46and keep the verdict `INCONCLUSIVE` unless the claim itself is static.4748For a matched installed-CLI example and verdict boundaries, read49[the synthetic example](references/synthetic-example.md).
SeasonCake