- Six declarative agents generated from the v1.1 stories (scaffold_from_stories.py, ADR 0001),
and a 9-step workflow export (build_workflow.py).
- Code-only intake (ADR 0008) verifies the one Discovery package in code before any agent.
- The rate card is the run input rate_card_document; no upload step (ADR 0009).
- Critic 0.1.6 after live runs on crucible-kab-dev:
- K-12 model independence checked at build time against each agent's deployed model
(ADR 0010); the critic reports it from its two inputs;
- receives in_cloud_estate for K-10; an unverified K-10 never fails the run alone;
- K-08 allows the scoring note to list findings as advisory, as ALZ-ASM-04 must;
- carries ALZ-ASM-03's and ALZ-ASM-04's rule sections, copied by the scaffold, so it can
recalculate risk, confidence and TCO.
- The scaffold refuses to regenerate unless every story file is present.
- Case azure-crucible-kab-dev-assessment with an illustrative rate card.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Case: azure-crucible-kab-dev-assessment
The first Assessment run on a live estate: the Discovery package of a run of case
azure-crucible-kab-dev-cmdb (resource group crucible-kab-dev, with the uploaded CMDB), and the illustrative
rate card written for this case (not real prices).
Starting it
From the repository root, after the Discovery run was continued past its review point:
F="individual-agents/ALZ-F02: Assessment & Disposition Workflow Agent"
python3 scripts/kab_run_workflow.py start --workflow <assessment workflow id> --project <id> \
--input engagement_id=ENG-AZURE-KAB-DEV-002 --input run_id=RUN-ASM-<n> \
--input discovery_execution_id=<discovery execution id> \
--input discovery_approval_ref="<who continued the Discovery run, and when>" \
--input 'artifact_ids=["<the package artifact id, <discovery execution id>_step_12>"]' \
--input rate_card_document=@"$F/workflows/cases/azure-crucible-kab-dev-assessment/rate-card.json"
Expected, written before the first run
From the Discovery package's facts and the stories, not from a run:
- Intake (
s0).verifiedis true: one package, every part matching its sha256, criticpass, from the named Discovery run. A package from any other run, or a changed one, fails the run here. - Applications. The planner plans one application, KAB Platform, with the package's 10 assets (F01 ADR 0021).
- Preconditions. Discovery completed, the critic passed and an approval reference was given.
- In-cloud estate. ALZ-ASM-05 (security posture) applies and reads the Azure estate with read-only tools.
- Risk. The end-of-life factor is missing for every asset (Discovery has no lifecycle source yet), so scores are
partialand nameeolinmissing_factors. No hard trigger can be evaluated. - TCO. Current cost is missing (Discovery reads no cost yet), so
current_annualis null and named in the missing inputs. The target uses this rate card. - Sign-off. Every disposition below 0.80 is in
signoff_queue, and so is every repurchase or relocate. The critic step stops for the Engagement Lead's approval. - Evidence. Every
evidence_refnames the Discovery execution id.