Files
gcp_solution_architecture_a…/providers/gcp.yaml
Jonathan Boniface 58e778ca7a feat: record verdicts in pieces and check the design against the schema
From the case 02 run cancelled at 66 minutes on 2026-10-07:

- The validator records verdicts a few Checks at a time (verdict_put) and
  they are attached on submission. It timed out five times writing them in
  one reply. Verdicts on the same code survive a run cut short; verdicts on
  changed code are dropped. Its model timeout is 300 s.
- A schema review at submission, in code: every setting a Check names is
  looked up on its Element's resource type, and an unknown resource type is
  reported. Two Revisions in that run were for Checks no code could meet.
  On the two designs that passed validation it raises 2 of 96 Checks, both
  one metric name used as a value, so it is returned once and never blocks.
- Reviews run together; the Jev coverage review stays off.

Plans record both steps and the cancelled run.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-08 09:39:13 +01:00

181 lines
6.3 KiB
YAML

# Cloud provider bundle: Google Cloud.
# Loaded by app/providers.py; select with CLOUD_PROVIDER=gcp. See ADR-0002.
name: gcp
skills_dirs:
- skills/vendor/google
- skills
output_template: skills/vendor/google/google-cloud-solution-architecture/assets/output-template.md
# Model ids: gemini/<model> runs on ADK's native Gemini client (Vertex AI or the Gemini
# API, chosen by environment); any other LiteLLM id runs through LiteLLM. Explicit
# versions, because -latest aliases are not served on Vertex AI (checked 2026-09-16).
# Reasoning-heavy roles use Pro; drafting and per-pillar advice use Flash.
models:
default: gemini/gemini-3.8-flash
roles:
orchestrator: gemini/gemini-3.1-pro-preview
product-selector: gemini/gemini-3.1-pro-preview
design-advisor: gemini/gemini-3.1-pro-preview
target-architect: gemini/gemini-3.1-pro-preview
iac-author: gemini/gemini-3.1-pro-preview
validator: gemini/gemini-3.1-pro-preview
# Seconds one model request may take, for roles whose replies are long. Other
# roles get MODEL_TIMEOUT_SECONDS. The target architect timed out at 180 s in both
# attempts of the 2026-10-07 run.
timeout_seconds:
target-architect: 420
iac-author: 300
validator: 300 # timed out five times at 180 s in a run of 2026-10-07
packager: 300
# Skills each role may load. Role files stay provider-neutral; the bundle names Skills.
role_skills:
orchestrator: [google-cloud-solution-architecture]
requirements-analyst: [google-cloud-solution-architecture, requirements-chain, cloud-discovery]
decomposer: [google-cloud-solution-architecture, requirements-chain]
product-selector: [google-cloud-solution-architecture, requirements-chain]
design-advisor: [google-cloud-solution-architecture, requirements-chain]
describer: [google-cloud-solution-architecture]
target-architect: [google-cloud-solution-architecture, requirements-chain]
iac-author: [google-cloud-solution-architecture, requirements-chain, iac-validation, google-cloud-iac-patterns]
validator: [google-cloud-solution-architecture, requirements-chain, iac-validation]
packager: [google-cloud-solution-architecture]
# One advisor per architecture-framework pillar; each loads only its pillar's Skill.
pillar_skills:
security: google-cloud-waf-security
reliability: google-cloud-waf-reliability
operational-excellence: google-cloud-waf-operational-excellence
cost-optimization: google-cloud-waf-cost-optimization
performance-optimization: google-cloud-waf-performance-optimization
sustainability: google-cloud-waf-sustainability
tools:
general: [mermaid_check, terraform_fmt, terraform_validate, terraform_schema]
cloud_read: [terraform_plan]
# The Terraform providers generated code is written for. One exact version, so the
# schema lookup, the patterns Skill and validation all describe the same provider.
# The 2026-10-07 run left five different versions in the plugin cache.
terraform:
required_providers:
google:
source: hashicorp/google
version: 8.6.0
mcp_auth:
kind: google_adc # Application Default Credentials bearer token; sends X-Goog-User-Project
# Remote MCP servers hosted by Google Cloud (streamable HTTP).
# `allow` lists name the only tools ever exposed. Reviewed 2026-09-16 against each
# server's live tools/list (scripts/list-mcp-tools.py): inventory and configuration
# metadata only, every tool readOnlyHint=true. Deliberately excluded even though
# read-only: data content (Cloud SQL execute_sql_readonly, Storage read_object/read_text,
# list_objects), generic Kubernetes reads that can return Secrets (get_k8s_resource,
# describe_k8s_resource), and every write. See docs/operations/mcp-access.md.
mcp_servers:
developer_knowledge:
url: https://developerknowledge.googleapis.com/mcp
tier: general # public documentation, no customer data
allow:
- search_documents
- get_documents
- answer_query
resource_manager:
url: https://cloudresourcemanager.googleapis.com/mcp
tier: cloud_read
allow:
- search_projects
asset_inventory:
url: https://cloudasset.googleapis.com/mcp
tier: cloud_read
allow:
- list_assets
- analyze_iam_policy
preview: true
compute:
url: https://compute.googleapis.com/mcp
tier: cloud_read
allow:
- list_instances
- get_instance_basic_info
- list_instance_attached_disks
- list_instance_group_managers
- get_instance_group_manager_basic_info
- list_instance_templates
- get_instance_template_properties
- list_disks
- get_disk_basic_info
- list_reservations
- list_commitments
gke:
url: https://container.googleapis.com/mcp
tier: cloud_read
allow:
- list_clusters
- get_cluster
- list_node_pools
- get_node_pool
- get_k8s_version
cloud_run:
url: https://run.googleapis.com/mcp
tier: cloud_read
allow:
- list_services
- get_service
cloud_sql:
url: https://sqladmin.googleapis.com/mcp
tier: cloud_read
allow:
- list_instances
- get_instance
storage:
url: https://storage.googleapis.com/storage/mcp
tier: cloud_read
allow:
- list_buckets
pubsub:
url: https://pubsub.googleapis.com/mcp
tier: cloud_read
allow:
- list_topics
- get_topic
- list_subscriptions
- get_subscription
iam:
url: https://iam.googleapis.com/mcp
tier: cloud_read
allow:
- list_roles
- get_role
- list_deny_policies
- get_deny_policy
monitoring:
url: https://monitoring.googleapis.com/mcp
tier: cloud_read
allow:
- list_alert_policies
- get_alert_policy
- list_dashboards
- list_metric_descriptors
- list_timeseries
network_intelligence:
url: https://networkmanagement.googleapis.com/mcp
tier: cloud_read
allow:
- list_connectivity_tests
- get_connectivity_test
preview: true
role_mcp_servers:
requirements-analyst:
[resource_manager, asset_inventory, compute, gke, cloud_run, cloud_sql, storage, pubsub, iam, monitoring]
product-selector: [developer_knowledge]
describer: [developer_knowledge]
design-advisor: [developer_knowledge]
target-architect: [developer_knowledge]
pillar-advisor: [developer_knowledge]
validator: [iam, network_intelligence]