Accounting problem
Before manager review, the preparer needs repeatable checks for completeness and unresolved blocking items against the configured rules.
Fictional inputs
The evaluation uses the repository fixtures examples/bas-not-ready and examples/bas-ready. They contain no client data.
Expected result
These results describe readiness for review; a manager still decides whether work may progress.
| Pack | Released expectation |
|---|---|
| Incomplete BAS | Exit 2 with NOT_READY. |
| Corrected BAS | Exit 0 with READY and no configured findings. |
Controls triggered
MISSING_ARTEFACTforgst_control_glSELF_REVIEW_INCOMPLETEOPEN_ITEM_BLOCKING
Human decision
READY means no configured gate tripped; it is not approval, advice or lodgement authority.
A manager still judges correctness, risk, materiality, client impact and whether the pack may progress.
Reproduce
git clone --branch review-ready-gate/v0.1.3 --depth 1 https://github.com/ryanduguid/accounting-review-pipeline.git
cd accounting-review-pipeline/packages/review-ready-gate
uv sync --locked --all-extras
uv run review-ready gate --profile bas --pack examples/bas-not-ready --output outputs/evaluation-not-ready
uv run review-ready gate --profile bas --pack examples/bas-ready --output outputs/evaluation-ready
uv run pytest tests/test_evaluation_pack.py -q
Primary sources
Versions
Product release v0.1.3; fixture version 1; source reviewed 2026-08-26.
review-ready-gate/v0.1.3 is the first release from the Accounting Review Pipeline monorepo. Its runtime behaviour and fictional fixtures are unchanged from v0.1.1, and the evaluation artefacts are fixed to that release’s commit.
Limitations
This evaluates configured gates against fictional inputs. It does not establish that a pack is correct, approve work, provide advice or authorise lodgement. See Evidence and Assurance for the common assurance record and boundaries.