open/portfolio-optimiser
1
0
Fork 0
Code Issues Releases 3 Activity
portfolio-optimiser/tests
History
Kjell Tore Guttormsen 688ee24973 feat(sim): the demo script is keyed on the project the prompt names, so a new project is data 
The scripted proposer answered one hard-coded pair of proposals. A second project meant a second
hand-written selector, written under demo-week time pressure -- the risk the week plan names
explicitly (§4, risk 2). It is now a registry: `ScriptedCandidate` entries selected by
`scripted_proposer`, plus `project_id` as an argument to `simulate_learning_loop`.

The open decision was WHAT identifies the candidate in the prompt blob; the plan flagged it as
unverified, so it was measured. Two prompt shapes reach the selector: the debate prompt carries the
whole bundle context, the generation prompt carries `Project: {id} - {name}` plus -- as its context
-- the debate output, which is the selector's own earlier reply. So the cost code and the measure
name are present in the generation prompt only because the script put them there; keying on them
would key the script on its own output. The project id is the one identifier both shapes carry and
the framework stamps.

Validation, never repair: no match, or more than one, raises `ScriptedCandidateError`. A default
reply would answer an unregistered project with another project's numbers, which on screen is
indistinguishable from a correct run; an ambiguous blob is a data problem that must surface at the
rehearsal rather than be decided by registry order.

Load-bearing MEASURED against the whole suite, five mutations all red plus a green control: detach
the project keying - one global flip key - fall back on an unknown project - first-match on an
ambiguous prompt - detach the `project_id` argument. The flip-key test was rewritten mid-measurement
because its first form asserted on the FIRST registry entry, where "the matched candidate's key" and
"candidates[0]'s key" coincide -- it could not separate the two implementations, and proved nothing.

766 passed / 4 skipped. Simulation still exits 0, still prints eight labelled steps, still
byte-identical across two runs.

[skip-docs] README is deliberately untouched: O4 defers the README rewrite to 14-15 August, after
the demo has produced the evidence for the level-2 claim. CLAUDE.md carries the invariant.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XoHJCKBTjFKcjsfEQyGbzh
2026-08-06 16:43:29 +02:00
..
spikes fix(fase3): relax hardcoded project-count assert in spike test (SC1 4th-project follow-up) 2026-06-26 12:14:34 +02:00
conftest.py feat(validator): anchor the deterministic gate to the project's real cost baseline (S4.0) 2026-08-03 17:19:31 +02:00
test_a5_per_approach_artifacts_loadbearing.py feat(outbox): every evaluated approach becomes something an expert can judge 2026-08-05 21:12:09 +02:00
test_assessment_method_loadbearing.py feat(fase1): assessment-method encoding — persisted dimension-scoped example + validator rule (F1) 2026-07-07 08:15:44 +02:00
test_b4_mcp_call_trace_loadbearing.py feat(provenance): a run records which external service it actually called 2026-08-05 21:37:29 +02:00
test_backends.py fix(s41): pass required credential in AzureFoundryBackend.create_chat_client 2026-07-15 11:18:40 +02:00
test_budget.py test(budget): defend the third field of the stop event — observed (kø-y) 2026-08-03 20:46:36 +02:00
test_bygg_energi_mikro.py docs(shared): bygg-energi mikro-eksempel — OKF-bundle + golden + load-bearing test 2026-06-29 09:42:13 +02:00
test_checker_gate_loadbearing.py feat(fase3): make the maker-checker checker actually gate the reasoning 2026-06-30 07:24:30 +02:00
test_content_keyed_script_loadbearing.py feat(sim): the demo script is keyed on the project the prompt names, so a new project is data 2026-08-06 16:43:29 +02:00
test_contracts.py feat(fase1): GoalContract + GoalConfig standalone fail-fast loader (F1) 2026-07-07 08:01:53 +02:00
test_costsim.py test(s36): value-level guard for _ore_to_kr_str money formatter 2026-07-15 10:25:05 +02:00
test_costsim_loadbearing.py feat(s36): costsim CLI + no-network + no-hardcoded-price guards 2026-07-15 10:05:29 +02:00
test_datasource.py feat(fase2): expose retriever as citation-bearing data source 2026-06-24 13:34:53 +02:00
test_dimension.py feat(s53): load_dimension fail-fast loader (MAF-free, mirrors load_goal_config) 2026-07-23 21:31:32 +02:00
test_dimension_loadbearing.py feat(fase1): dimension in run_project — context scope + candidate constraint (F1) 2026-07-07 07:50:26 +02:00
test_doc_constant_sync_loadbearing.py docs(mandate): how a domain expert commissions a run — and one honesty fix the run itself exposed 2026-08-05 16:31:27 +02:00
test_foundry_profile_live.py test(fase2): gated live local + Foundry profile checks 2026-06-24 14:02:44 +02:00
test_frontmatter_unquote_loadbearing.py fix(okf): one unquoting rule for frontmatter scalars (kø-(a)/(i)) 2026-08-04 21:23:18 +02:00
test_generate.py feat(step5): the falsification that informed the next hypothesis now leaves the loop 2026-08-06 15:12:06 +02:00
test_hitl.py fix(s51): close 2 review MINORs — non-UTF-8 skip + affected_codes parity 2026-07-15 20:59:01 +02:00
test_hitl_loadbearing.py fix(s51): close 2 review MINORs — non-UTF-8 skip + affected_codes parity 2026-07-15 20:59:01 +02:00
test_imports_core.py feat(fase1): export domain-model public API + green full suite (F1) 2026-07-07 08:17:10 +02:00
test_ingest_golden.py test(ingest): golden extraction case examples/ingest-golden-file, byte-frozen (I2) 2026-07-03 18:40:53 +02:00
test_ingest_golden_http.py chore(repo): planning artifacts become local-only; fixture builders become code 2026-08-05 10:08:17 +02:00
test_ingest_golden_mcp.py fix(ingest): narrow mcp_timeout to our OWN deadline, not the exception type (kø-z follow-up) 2026-08-03 21:22:00 +02:00
test_ingest_golden_sql.py chore(repo): planning artifacts become local-only; fixture builders become code 2026-08-05 10:08:17 +02:00
test_ingest_http.py feat(ingest): http read_http connector + injectable transport seam (I6) 2026-07-04 16:57:29 +02:00
test_ingest_http_loadbearing.py feat(ingest): gate http behind allow_network run-flag + dispatch (I6) 2026-07-04 17:04:05 +02:00
test_ingest_http_timeout_loadbearing.py feat(ingest): bound the default http transport in time, in front of the pinned library (S2.4) 2026-08-03 14:54:04 +02:00
test_ingest_library_seam.py chore(deps): move the ingest library pin to v0.3.2 — as far as latest goes today 2026-08-05 12:11:12 +02:00
test_ingest_loadbearing.py fix(tests): the re-ingest fixture modelled two manifests, not one edited manifest 2026-08-05 12:30:08 +02:00
test_ingest_manifest.py refactor(ingest): adopt shared llm-ingestion-okf v0.3.1 behind a thin adapter 2026-07-20 07:47:55 +02:00
test_ingest_materialize.py refactor(ingest): adopt shared llm-ingestion-okf v0.3.1 behind a thin adapter 2026-07-20 07:47:55 +02:00
test_ingest_mcp.py chore(deps): move the ingest library pin to v0.3.2 — as far as latest goes today 2026-08-05 12:11:12 +02:00
test_ingest_mcp_loadbearing.py feat(ingest): add the MCP connector as a transport inside the http family (S2.2) 2026-08-02 21:14:57 +02:00
test_ingest_sql.py test(ingest): SQL connector + load-bearing + golden, detach-proven (I4) 2026-07-04 06:56:02 +02:00
test_ingest_sql_loadbearing.py test(ingest): SQL connector + load-bearing + golden, detach-proven (I4) 2026-07-04 06:56:02 +02:00
test_ledger.py fix(s54): reject non-array/non-object ledger JSON in SavingsLedger.load 2026-07-24 19:52:52 +02:00
test_ledger_loadbearing.py feat(fase1): hard/soft goal-stop in run_portfolio on accumulated ledger (F1) 2026-07-07 08:11:43 +02:00
test_live_dry_run.py fix(s42): close review WARN — hermetic dry-run env + scoped azure hint 2026-07-15 18:30:30 +02:00
test_live_dry_run_loadbearing.py feat(s42): live_dry_run cut in run_project + DryRunReport 2026-07-15 18:05:16 +02:00
test_local_profile_live.py test(fase2): gated live local + Foundry profile checks 2026-06-24 14:02:44 +02:00
test_maf_version_guard.py feat(fase2a): MAF version-guard (test-tidsvakt, dist-metadata) + pin agent-framework-core<2 (S2.5) 2026-07-15 07:28:40 +02:00
test_mandate.py docs(mandate): how a domain expert commissions a run — and one honesty fix the run itself exposed 2026-08-05 16:31:27 +02:00
test_mandate_cli.py feat(mcp): concrete MCP servers become tools the agents can call during a run 2026-08-05 16:53:07 +02:00
test_mandate_generation_loadbearing.py feat(step5): the falsification that informed the next hypothesis now leaves the loop 2026-08-06 15:12:06 +02:00
test_mandate_run_loadbearing.py feat(run): every commissioned approach is evaluated, and every one is reported 2026-08-05 16:05:48 +02:00
test_mcp_run_loadbearing.py feat(mcp): concrete MCP servers become tools the agents can call during a run 2026-08-05 16:53:07 +02:00
test_mcp_tools.py feat(mcp): concrete MCP servers become tools the agents can call during a run 2026-08-05 16:53:07 +02:00
test_method_spec_loadbearing.py test(spec): make the pulled method spec load-bearing at rule level, not just section level 2026-08-01 19:57:33 +02:00
test_money_quantization_loadbearing.py fix(money): quantize NOK to øre in one order, from one source (kø-p) 2026-08-03 20:08:59 +02:00
test_notify.py test(s52): cover build_notifier console/file dispatch + file-without-path fail-fast 2026-07-17 03:16:25 +02:00
test_notify_loadbearing.py fix(s52): resolve ConsoleNotifier default stream at call time 2026-07-17 03:18:00 +02:00
test_okf.py feat(mandate): a domain expert can commission WHICH approaches a run evaluates 2026-08-05 15:43:02 +02:00
test_outbox_loadbearing.py feat(s42): byte-deterministic write_run_config outbox writer 2026-07-15 12:39:42 +02:00
test_persona_skill_loadbearing.py feat(persona): build the shared expert-reviewer persona as a framework-neutral Agent Skill 2026-06-30 13:59:42 +02:00
test_portfolio.py feat(validator): anchor the deterministic gate to the project's real cost baseline (S4.0) 2026-08-03 17:19:31 +02:00
test_portfolio_budget_loadbearing.py feat(validator): anchor the deterministic gate to the project's real cost baseline (S4.0) 2026-08-03 17:19:31 +02:00
test_portfolio_cli_offline_loadbearing.py fix(tests): blade 2 was testing the wrong defect, and said so out loud 2026-08-05 11:06:41 +02:00
test_portfolio_concurrent_loadbearing.py test(portfolio): gate the wave handler's catch width and the failed-project ledger (v/t/s) 2026-08-03 16:09:12 +02:00
test_portfolio_failure_accounting_loadbearing.py feat(validator): anchor the deterministic gate to the project's real cost baseline (S4.0) 2026-08-03 17:19:31 +02:00
test_portfolio_learning_loadbearing.py feat(fase2a): percent-mål mot baseline 0 reiser ValueError (S2.0) 2026-07-15 07:14:29 +02:00
test_portfolio_live.py test(fase3): SC7 both profiles offline + gated Azure portfolio arm 2026-06-26 12:10:04 +02:00
test_preflight.py fix(s41): close review WARN — OSError refusal + scheme/profile branch coverage 2026-07-15 11:58:10 +02:00
test_preflight_loadbearing.py style(s41): ruff format preflight test files 2026-07-15 11:25:26 +02:00
test_provenance.py feat(fase2): first-class Pydantic provenance stamp 2026-06-24 13:26:55 +02:00
test_reference_domain.py test(fase2a): commit Step 2 optional bundle_dir/verdict_dir tests (S2.0, omitted from 9659045) 2026-07-15 07:29:19 +02:00
test_retrieval.py fix(retrieval): fail-closed on uncanonicalisable paths (embedded null byte) 2026-07-21 08:00:11 +02:00
test_run_cli.py fix(run): --embedder-config is refused, not silently dropped 2026-08-05 11:39:49 +02:00
test_run_cli_loadbearing.py test(s31): close 1 review MAJOR — main()-level load-bearing proof of the --semantic-retrieval wiring 2026-07-25 12:44:03 +02:00
test_run_smoke.py feat(fase2): vertical-slice orchestrator + two-layer HITL wiring 2026-06-24 13:54:54 +02:00
test_s40_cost_baseline_loadbearing.py feat(validator): anchor the deterministic gate to the project's real cost baseline (S4.0) 2026-08-03 17:19:31 +02:00
test_scripted_cli_door_loadbearing.py feat(run): a CLI door onto the offline whole-loop run (--scripted-replies) 2026-08-05 08:55:37 +02:00
test_scripted_client_consolidation.py feat(provenance): a run records which external service it actually called 2026-08-05 21:37:29 +02:00
test_semretrieval.py fix(semretrieval): refuse a non-finite embedding instead of scoring it (kø-(l)/S3.1 MINOR) 2026-08-03 21:48:50 +02:00
test_semretrieval_loadbearing.py fix(semretrieval): refuse a non-finite embedding instead of scoring it (kø-(l)/S3.1 MINOR) 2026-08-03 21:48:50 +02:00
test_shared_root_loadbearing.py feat(shared-root): S3 — configurable shared-root resolver with load-bearing override test 2026-07-03 01:10:05 +02:00
test_simulation_loadbearing.py feat(sim): offline end-to-end simulation proving the learning loop closes 2026-06-30 12:55:15 +02:00
test_smoke.py feat: initial scaffold (Python framework on Microsoft Agent Framework) 2026-06-23 22:01:22 +02:00
test_step1_expel_loadbearing.py feat(fase2b): OKF-navigated bundle context replaces chunk-stuffing 2026-06-30 06:42:19 +02:00
test_step5_history_loadbearing.py feat(step5): the falsification that informed the next hypothesis now leaves the loop 2026-08-06 15:12:06 +02:00
test_step5_refine_loadbearing.py feat(step5): the falsification that informed the next hypothesis now leaves the loop 2026-08-06 15:12:06 +02:00
test_step7_async_loop_loadbearing.py feat(fase5): add the long/async verdict file inbox (Steg 7 resumable feedback) 2026-06-30 09:54:23 +02:00
test_step8_promotion_loadbearing.py feat(verdicts): key each verdict on its own candidate, not the bundle's one IR projection (S3.2) 2026-08-03 16:44:58 +02:00
test_step32_multicandidate_loadbearing.py feat(verdicts): key each verdict on its own candidate, not the bundle's one IR projection (S3.2) 2026-08-03 16:44:58 +02:00
test_validator.py feat(validator): enforce the deterministic stage-2 bound and band enclosure (S2.7) 2026-08-03 16:22:54 +02:00
test_value_report.py feat(s54): value_report text + deterministic JSON formatters 2026-07-24 01:29:31 +02:00
test_value_report_loadbearing.py test(s54): value_report overlap-flag load-bearing seam 2026-07-24 01:30:55 +02:00
test_verdicts.py fix(s51): close 2 review MINORs — non-UTF-8 skip + affected_codes parity 2026-07-15 20:59:01 +02:00
test_vertical_slice_e2e.py fix(fase3): stamp real proposer model into provenance, kill fake-model leak (F1) 2026-06-26 15:03:45 +02:00
test_workflow.py feat(fase2): thread tools + budget middleware through fresh_workflow onto agents 2026-06-26 00:36:06 +02:00