Correcting bot commit

Author: Deoncyril

.github/workflows/integration-tests.yml

@@ -0,0 +1,38 @@
+name: Integration Tests
+
+on:
+  workflow_dispatch:
+
+jobs:
+  integration-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+      - name: Install system deps (docker-compose)
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y docker.io docker-compose-plugin
+      - name: Install Poetry
+        run: pip install poetry
+      - name: Install dependencies
+        run: poetry install --with dev
+      - name: Start integration services
+        run: |
+          docker compose -f tests/integration/docker-compose.yml up -d
+      - name: Wait for services
+        run: |
+          for i in {1..30}; do
+            if curl -sS http://localhost:9200/ >/dev/null 2>&1; then break; fi
+            sleep 1
+          done
+      - name: Run integration tests
+        env:
+          RUN_INTEGRATION: '1'
+        run: poetry run pytest -q tests/integration -m integration
+      - name: Tear down services
+        if: always()
+        run: docker compose -f tests/integration/docker-compose.yml down -v

README.md

@@ -4,9 +4,6 @@ Skeleton project for RAG/LLM orchestration inspired by the anomaly-detector impl
 
 Quick start
 
-- Install dependencies: poetry install
-- Run tests: poetry run pytest -q
-- To run the PaS estimator locally:
   - poetry run python tools/pas_estimate.py
 
 Local development (quick)
@@ -48,13 +45,6 @@ python scripts/update_checklist.py --file checklist.md --index 3
 
 Project structure
 
-- infra/: CDK stack and infra code
-- src/rag_api/: FastAPI application and retriever code
-- src/orchestrator/: Step Functions / workflow skeleton
-- config/: environment YAMLs
-- scripts/: deployment and packaging scripts (dry-run by default)
-- tools/: PaS estimator and CI helpers
-- tests/: unit tests
 
 Docker
 
@@ -69,3 +59,13 @@ Run container (example):
 ```bash
 docker run --rm -p 8000:8000 -e AWS_REGION=us-west-2 -e VECTOR_STORE_TYPE=opensearch answer-architect:local
 ```
+
+Testing strategy
+---------------
+
+Unit tests in `tests/` are designed to run quickly and use lightweight fakes for heavy external
+dependencies (OpenSearch, boto3, SentenceTransformers). Integration tests that exercise real
+services should be placed under `tests/integration/` and run in a CI job or developer environment
+that installs the real dependencies.
+
+See `docs/TESTING.md` for more details.

docs/CI.md

@@ -0,0 +1,27 @@
+CI runbook - Answer Architect
+
+This document describes the repository CI workflows and how to safely enable deploys.
+
+Workflows included in this repo
+- CI - tests and update checklist (`.github/workflows/update-checklist.yml`)
+  - Runs on push and pull_request to `main`/`master`.
+  - Installs dependencies (Poetry), runs unit tests, runs `python scripts/update_checklist.py --recalc` and commits `checklist.md` when changed.
+  - Does not perform any cloud deploys.
+
+- Manual Deploy (safe) (`.github/workflows/manual-deploy.yml`)
+  - Manual `workflow_dispatch` only. Requires typing the confirmation phrase `I_ACCEPT_COSTS` to run preflight checks.
+  - Intentionally performs a dry-run and refuses to `apply` from Actions unless repository administrators explicitly enable CI deploy flags and environment approvals.
+
+How to enable safe CI deploys (admin steps)
+1. Create a GitHub Environment (Settings -> Environments) with the name you will use (e.g., `production`).
+2. Configure required reviewers or environment protection rules so deployments to that environment require human approval.
+3. If you understand the risk and want to allow the apply step to run from Actions, set the repository secrets or environment variables (in the environment settings) `ALLOW_DEPLOY_IN_CI=1` and `ALLOW_AWS_DEPLOY=1`.
+   - Warning: this is a high-risk change. Only set these when environment approvals are configured and reviewers are trained.
+
+Local safe deploy checklist
+- Create a `.deploy-confirm` file containing `I_ACCEPT_COSTS` if you want non-interactive confirmation.
+- Set `ALLOW_AWS_DEPLOY=1` and `DEPLOY_CONFIRM=I_ACCEPT_COSTS` in your shell, then run `python scripts/deploy.py --environment dev --apply`.
+
+Notes
+- The CI runbook is informational; the repository includes guardrails in `scripts/deploy.py` that block accidental deploys.
+- If you want me to generate admin-facing automation (GH CLI scripts) to create environments and apply protection rules, I can create them, but they require repository admin tokens to run.

docs/DEPLOY.md

@@ -1,4 +1,4 @@
-# Safe deploy runbook — Answer Architect
+# Safe deploy runbook - Answer Architect
 
 This document describes the safe, audited steps to deploy the Answer Architect infrastructure.
 

docs/ENV_SETUP.md

@@ -0,0 +1,30 @@
+# Environment setup
+
+Quick steps to prepare a developer workstation for Answer Architect:
+
+1. Install Python 3.11. Example with pyenv:
+
+```bash
+pyenv install 3.11.4
+pyenv local 3.11.4
+```
+
+2. Install Poetry and project deps:
+
+```bash
+curl -sSL https://install.python-poetry.org | python3 -
+poetry install
+```
+
+3. (Optional) Install AWS CLI v2 and configure profile for managed LLM access if needed:
+
+```bash
+# follow AWS docs for your OS; verify:
+aws --version
+```
+
+4. Quick check:
+
+```bash
+python scripts/check_env.py
+```

docs/TESTING.md

@@ -0,0 +1,35 @@
+Testing strategy for Answer Architect
+
+Unit tests
+----------
+- Fast.
+- Live under `tests/`.
+- Use `tests/conftest.py` to inject lightweight fakes for heavy external dependencies (boto3, opensearchpy, sentence_transformers, pgvector/psycopg2, aws_requests_auth, tiktoken).
+- Keep unit tests focused on logic and structure, not external service behavior.
+
+Integration tests
+-----------------
+- Place under `tests/integration/`.
+- These should be run in an environment that installs the real dependencies (e.g., via Poetry or a reproducible Docker image).
+- Integration tests may require running local or cloud services (OpenSearch, PostgreSQL with pgvector, Bedrock-mock) or using testcontainers to spin up required services.
+
+Running tests locally
+---------------------
+- Run unit tests quickly:
+
+```bash
+poetry install --with dev
+poetry run pytest tests/ -q
+```
+
+- Run integration tests (example, requires real services):
+
+```bash
+poetry install --with dev
+poetry run pytest tests/integration/ -q
+```
+
+CI
+--
+- Keep unit tests fast and run them on every PR.
+- Run integration tests in a separate job that provisions required services and installs full dependencies.

pytest.ini

@@ -0,0 +1,5 @@
+[pytest]
+markers =
+    asyncio: mark async tests (pytest-asyncio)
+    latency: mark tests that assert latency budgets
+    integration: mark tests that require external services (OpenSearch/Postgres)

scripts/__pycache__/update_checklist.cpython-311.pyc

@@ -0,0 +1,58 @@
+#!/usr/bin/env python3
+"""Check developer environment for required toolchain versions.
+
+Usage: python scripts/check_env.py
+Exits 0 if checks pass, non-zero otherwise.
+"""
+from __future__ import annotations
+
+import shutil
+import subprocess
+import sys
+
+
+def check_python(min_major: int = 3, min_minor: int = 11) -> bool:
+    v = sys.version_info
+    ok = (v.major > min_major) or (v.major == min_major and v.minor >= min_minor)
+    print(f"Python: {v.major}.{v.minor}.{v.micro} -> {'OK' if ok else 'FAIL (need >=%d.%d)' % (min_major, min_minor)}")
+    return ok
+
+
+def check_executable(name: str) -> bool:
+    path = shutil.which(name)
+    print(f"{name}: {'found at ' + path if path else 'NOT FOUND'}")
+    return bool(path)
+
+
+def check_aws_cli(min_version: tuple[int, int, int] = (2, 27, 50)) -> bool:
+    if not check_executable("aws"):
+        return False
+    try:
+        out = subprocess.check_output(["aws", "--version"], stderr=subprocess.STDOUT)
+        s = out.decode("utf-8", errors="ignore")
+        # Example: aws-cli/2.7.18 Python/3.9.16 Linux/5.15.0
+        parts = s.split()[0].split("/")
+        if len(parts) >= 2:
+            ver = parts[1]
+            nums = tuple(int(p) for p in ver.split(".")[:3])
+            ok = nums >= min_version
+            print(f"AWS CLI: {ver} -> {'OK' if ok else 'FAIL (need >=%s)' % ('.'.join(map(str,min_version)))}")
+            return ok
+    except Exception as e:
+        print("Error checking aws --version:", e)
+    return False
+
+
+def main() -> None:
+    ok = True
+    ok = check_python() and ok
+    ok = check_executable("poetry") and ok
+    ok = check_aws_cli() and ok
+    if not ok:
+        print("One or more environment checks failed.")
+        sys.exit(2)
+    print("Environment OK")
+
+
+if __name__ == "__main__":
+    main()