> Data revision: baf7c2f4307d3a989a8fb9363df8000fda3d9310
> Source files:
- [EVIDENCE-MAP.md](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/EVIDENCE-MAP.md)

# 2.4 Acceptance tests and stack adapters

> The clause text and evidence status below are from the source map. Numeric figures remain source-reported unless tied to a claim-ledger record.

**Map clause:** Acceptance ledger, adapters, and findings.

- **Source heading anchors:** [spec/02-formats.md#24-acceptance-tests-and-stack-adapters](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#24-acceptance-tests-and-stack-adapters); [spec/02-formats.md#241-contract](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#241-contract); [spec/02-formats.md#242-adapter-interface](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#242-adapter-interface); [spec/02-formats.md#243-built-in-adapters](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#243-built-in-adapters); [spec/02-formats.md#244-findings-and-identities](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#244-findings-and-identities); [spec/02-formats.md#245-shape-artefacts](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/spec/02-formats.md#245-shape-artefacts)

- **Spec revision / decision reference:** v1.2 (pinned fragment locally verified); section-level D references are listed in the source classification. **Last reviewed:** 2026-10-07.
- **Type:** Safety/correctness for ledger interpretation and adapter behavior; empirical for detection and repair benefit.
- **Evidence and status:** [R55](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/rounds/r55/README.md) and [R70](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/rounds/r70/README.md) are **INTERIM** for the questions needed here. The Intent coverage, over-strict replay, and candidate counterfactual records are offline diagnostics: `CL-WA-intent-coverage-audit-reported-observation` (**INTERIM**), `CL-WA-overstrict-replay-reported-observation` (**INTERIM**), and `CL-WA-candidate-counterfactual-reported-observation` (**INTERIM**). No prospective live repair effect is established. Related hypotheses: H62–H74, H80, H89, H142.
- **Unresolved measurement:** Test adapter parity, true-defect detection, false blocks, and whether a repair loop improves official task outcomes on a prospective cohort.

## Related records

### Related hypotheses

- [H62](/hypotheses/h62/)
- [H74](/hypotheses/h74/)
- [H80](/hypotheses/h80/)
- [H89](/hypotheses/h89/)
- [H142](/hypotheses/h142/)

### Related experiments

- [r55](/rounds/r55/)
- [r70](/rounds/r70/)

### Related claims

- [CL-WA-candidate-counterfactual-reported-observation](/claims/cl-wa-candidate-counterfactual-reported-observation/)
- [CL-WA-intent-coverage-audit-reported-observation](/claims/cl-wa-intent-coverage-audit-reported-observation/)
- [CL-WA-overstrict-replay-reported-observation](/claims/cl-wa-overstrict-replay-reported-observation/)
