Files
cp1-agents/individual-agents/ALZ-F02: Assessment & Disposition Workflow Agent/CLAUDE.md
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

2.6 KiB

Individual agents: ALZ-F02 Assessment & Disposition

CP1 Assessment as six separate declarative KAB agents chained by the workflow, cp4-agents style, behind a code-only intake step (ADR 0008). Its counterpart, ../../sub-agents/ALZ-F02: Assessment & Disposition Workflow Agent/, is the sub-agent pattern (not built yet). Keep the two patterns apart: never copy app/, sub-agents or workspace code into this folder, and never merge their ADRs.

@.agents/CONTEXT.md @.agents/PLAN.md

Each agent folder has its own CLAUDE.md and .agents/CONTEXT.md. Repository-wide rules are in ../../CLAUDE.md.

Commands (from this folder)

python3 scripts/scaffold_from_stories.py   # regenerate the six agent folders from the v1.1 story files
python3 scripts/build_workflow.py          # after any schemas/ change; it asserts every from_steps field exists
python3 ../../scripts/check_agents.py --runtime-image cruciblecoredevacr.azurecr.io/kab/agent-runtime-adk:0.35.7 cp1-*/
(cd ../.. && python3 scripts/render_k8s.py deploy/crucible-kab-dev.yaml <agent> && kustomize build build/crucible-kab-dev/<agent>)

Rules

  • Declarative agents only: no app/, functions or hooks. Scoring, confidence and TCO are prompt rules that the critic recalculates (ADR 0001). Don't add a scoring tool or an MCP server for them. The one exception is cp1-assessment-intake-agent (ADR 0008): code only, not generated by the scaffold, and it only checks the Discovery package. Keep scoring out of it.
  • Existing MCP servers only. ALZ-ASM-02 and ALZ-ASM-05 use azure-discovery, aws-discovery and gcp-discovery. ALZ-ASM-05's guidance comes from <provider>_search_well_architected on the same servers (Azure and GCP; none on AWS yet). Read-only tools only.
  • The agent folders are generated by scripts/scaffold_from_stories.py from the story files in the engineering hub workspace. Change the script and regenerate; regenerating rewrites whole folders, so review the diff.
  • Discovery's outputs arrive as one package, as text, through a1 (Artifacts) and p1 (parse) (ADRs 0002, 0008), and the intake s0 checks it before the planner runs. Never wire an agent to a1 directly: its only output, artifacts, holds base64 records.
  • Keep repo: cp1-agents-individual on these agents in ../../deploy/.
  • A change to an original requirement needs an ADR in docs/adr/, a link to it from each affected agent's docs/adr/README.md (in ADRS_FOR in the scaffold), and a step in .agents/PLAN.md.
  • No pattern detail on ADO tickets. Record pattern decisions here, not on the board.