Files
Jonathan Boniface e22d037ab0 ALZ-F02 Assessment: individual agents, intake, rate card input, critic fixes
- 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>
2026-11-05 22:13:25 +00:00
..

Workflow cases: ALZ-F02 Assessment

Case Input Status
azure-crucible-kab-dev-assessment The Discovery package of a live Azure run with an uploaded CMDB (F01 ADRs 0020, 0021), plus an illustrative Azure rate card Expected behaviour written before the first run
meridian-assessment A Meridian Retail Group Discovery run with a passing critic and an approval, plus the sample rate card Rate card only. The checks are written before the first run (.agents/PLAN.md step 10)

Expected behaviour on Meridian, from the story files' worked examples. Checks should be written from this list before the first run, not tuned to it afterwards:

  • No cloud. The estate is on-premises, so ALZ-ASM-05 is not_applicable and makes no tool call.
  • Risk. POS Backoffice is in the high-risk tail: its score is 0.85, and it meets both hard triggers.
  • Partial score. The file-transfer score is partial, and T1 can't be evaluated because its criticality isn't recorded.
  • Dependents. No disposition retires an application with a dependent, and none retires an asset whose retire_constraint isn't none.
  • Sign-off. Every disposition below 0.80 is in signoff_queue, and so is every repurchase or relocate.
  • Evidence. Every evidence_ref names the Discovery run's execution id.