The reference agent's subject, cloud solution architecture, as an agent in the operating model agent's pattern: the Execution engine (from orchestration_engine) and one Workflow in app/solution_architecture, built for the settings and provider bundle by assembly.workflow_for. Copied from gcp_solution_architecture_agent at 8b397cd without its own engine: its role files, Skills, provider bundle, ADRs, vocabulary, scripts and evaluation datasets. The image takes the engine from a named build context and installs Terraform. It is called through the Orchestration engine, and its files are what the operating model agent digests. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
1.4 KiB
CLAUDE.md
Direction
This agent turns a request for a cloud solution into a reviewed architecture, its infrastructure code and a plan to build it. It is the Execution engine (../orchestration_engine) and one Workflow, app/solution_architecture/workflow.py, assembled with the settings and provider bundle by app/solution_architecture/assembly.py. Read CONTEXT.md and docs/adr/ before designing or restructuring anything.
It is called through the Orchestration engine (../orchestration_engine/orchestration/), which every request comes through. Its files are what the operating model agent (../tom_analysis) digests into an operating spec, so a change to their format is a change to that agent's digest.
The reference it was built from, ../gcp_solution_architecture_agent, is only ever read. The ADRs in docs/adr/ were written there and name its module paths; app/solution_architecture/ holds the subject's modules under the same names.
Engine code is not edited here. A fault in it is fixed in ../orchestration_engine.
Plans and history live in .agents/plans/.
Checks
A step is done when all three pass:
.venv/bin/python -m pytest -q
.venv/bin/ruff check . && .venv/bin/ruff format --check .
.venv/bin/mypy .
Set up with pip install -e ../orchestration_engine --config-settings editable_mode=compat and then pip install -e '.[dev]'.