Inspect App

.claude/rules/agent-rules-layout.md

@@ -0,0 +1,67 @@
+---
+description: Agent rules directory layout and symlink conventions
+alwaysApply: false
+globs: .claude/rules/**,.cursor/rules/**
+paths:
+  - ".claude/rules/**/*.md"
+  - ".cursor/rules/**/*.mdc"
+---
+
+# Agent rules layout
+
+This repo shares agent rules between **Claude Code** and **Cursor** via symlinks.
+
+## Canonical source
+
+- **Edit rules here:** `.claude/rules/*.md`
+- **Do not** put rule content directly in `.cursor/rules/` — those files are symlinks.
+
+## Cursor bridge
+
+Cursor requires `.mdc` files in `.cursor/rules/`. Each `.mdc` is a symlink to the matching `.md`:
+
+```text
+.claude/rules/my-rule.md          ← canonical (commit this)
+.cursor/rules/my-rule.mdc         ← symlink → ../.claude/rules/my-rule.md
+```
+
+## Dual frontmatter
+
+Every rule file needs frontmatter for both tools:
+
+```yaml
+---
+description: Short summary for Cursor rule picker
+alwaysApply: false
+globs: path/to/match/**
+paths:
+  - "path/to/match/**"
+---
+```
+
+- **Cursor** uses `globs`, `alwaysApply`, and `description`.
+- **Claude Code** uses `paths`; omit `paths` for always-on rules.
+- Each tool ignores the other's keys.
+
+## Adding a rule
+
+1. Create `.claude/rules/<name>.md` with dual frontmatter and content.
+2. Add the Cursor symlink:
+
+```bash
+ln -s ../.claude/rules/<name>.md .cursor/rules/<name>.mdc
+```
+
+3. List the new rule in the **Agent rules** section of `AGENTS.md`.
+
+## Removing a rule
+
+1. Delete `.claude/rules/<name>.md`.
+2. Delete `.cursor/rules/<name>.mdc` (the symlink, not a separate file).
+3. Remove its entry from `AGENTS.md`.
+
+## Do not
+
+- Symlink the entire `.cursor/rules` directory to `.claude/rules` (extension mismatch: `.mdc` vs `.md`).
+- Duplicate rule content in both directories.
+- Commit real files under `.cursor/rules/` — only symlinks belong there.

.claude/rules/python-quality.md

@@ -0,0 +1,46 @@
+---
+description: Python code quality conventions for src/ and tests/
+alwaysApply: false
+globs: src/**/*.py,tests/**/*.py
+paths:
+  - "src/**/*.py"
+  - "tests/**/*.py"
+---
+
+# Python code quality
+
+## Structured data
+
+- Use **Pydantic `BaseModel`** for all structured data — API responses, settings, internal records, result types, and domain wrappers. Do not use `@dataclass`.
+- Prefer typed models over plain `dict` for structured data.
+- Reuse project base models where they exist (e.g. `InspectApiBaseModel` for API shapes; inspect-internal bases for in-memory models).
+
+## Exceptions
+
+- Raise **specific, descriptive exceptions** rather than bare `Exception`.
+- Follow domain error hierarchies: a base exception class plus typed subclasses with context attributes (see `src/videoipath_automation_tool/apps/inspect/errors.py`).
+
+## Context managers
+
+- Use `with` for files, locks, test spies, and connector wrappers.
+
+## Quality gates
+
+Before finishing Python changes, run:
+
+```bash
+poetry run ruff check --fix src/ tests/
+poetry run ruff format src/ tests/
+```
+
+Pre-commit hooks mirror these commands (see `.pre-commit-config.yaml`).
+
+## Generated and sensitive code
+
+- Do **not** hand-edit `src/videoipath_automation_tool/apps/inventory/model/drivers.py` — regenerate with `set-videoipath-version <version>`.
+- All committed data must follow the anonymization rules in `AGENTS.md`.
+
+## Change scope
+
+- Keep diffs minimal and focused.
+- Match surrounding patterns: mixins, module docstrings, `TYPE_CHECKING` imports, and existing naming in the file you edit.

.claude/rules/python-style.md

@@ -0,0 +1,55 @@
+---
+description: Python coding style for src/ and tests/
+alwaysApply: false
+globs: src/**/*.py,tests/**/*.py
+paths:
+  - "src/**/*.py"
+  - "tests/**/*.py"
+---
+
+# Python coding style
+
+## Type hints
+
+- Annotate all function parameters and return types.
+- Use `from __future__ import annotations` in new modules.
+- Use `TYPE_CHECKING` blocks for imports needed only for type hints.
+
+## Formatting and naming
+
+- Follow PEP 8 with **snake_case** for functions, variables, and modules.
+- Max line length is **120** characters (ruff formatter default in this project — not Black's 88).
+- Format with **ruff**, not Black:
+
+```bash
+poetry run ruff format src/ tests/
+```
+
+## Strings and paths
+
+- Use **f-strings** for string formatting; avoid `%` formatting and `.format()`.
+- Use **`pathlib.Path`** over `os.path` for filesystem operations.
+
+## Comprehensions and readability
+
+- Prefer list/dict/set comprehensions over explicit loops when the result stays readable.
+- Do not sacrifice clarity for brevity.
+
+## Layout and whitespace
+
+- Group **logically related lines** together (e.g. setup, core logic, cleanup).
+- Separate groups with a **single blank line**; use an extra blank line between larger sections when it aids scanning.
+- Do not sprinkle blank lines randomly, and do not leave long unbroken blocks when a visual break would help.
+- Within a function, keep the main path easy to follow: inputs and validation first, then the core work, then return/cleanup.
+
+## Public before private
+
+- Place **public** API first so readers see the most relevant surface when scrolling: public classes, methods, functions, and module-level constants.
+- Place **private** members after public ones: names prefixed with `_` (attributes, methods, functions, nested helpers) and internal implementation details.
+- In classes: public methods first, then `_`-prefixed helpers and internal state accessors.
+- In modules: public exports and user-facing functions first; private helpers and module-internal constants at the bottom.
+- A short section comment (e.g. `# --- Internal ---`) is fine when a class or module has a large private block.
+
+## Resources
+
+- Use **context managers** (`with`) for files, locks, and other resources that need cleanup.

.claude/rules/python-testing.md

@@ -0,0 +1,81 @@
+---
+description: Python testing conventions for src/ and tests/
+alwaysApply: false
+globs: src/**/*.py,tests/**/*.py
+paths:
+  - "src/**/*.py"
+  - "tests/**/*.py"
+---
+
+# Python testing
+
+## Framework and commands
+
+- Use **pytest** for all tests.
+- Prefer the dedicated entry points over bare `pytest`:
+
+```bash
+poetry run test-unit              # offline/unit suite (CI default)
+poetry run test-e2e               # live-server e2e (loads .env)
+poetry run test               # unit then e2e sequentially
+
+# Single file or test (extra args pass through to test-unit / test-e2e)
+poetry run test-unit tests/validators/test_device_id.py
+poetry run test-e2e tests/e2e/inspect/test_e2e_inspect.py::test_name
+```
+
+- `poetry run pytest` also runs unit tests only (e2e excluded via `addopts` in `pyproject.toml`).
+
+### VS Code
+
+Use the launch configs in `.vscode/launch.json`:
+
+- **Unit Tests** / **Unit Tests (current file)** — offline suite
+- **E2E Tests** / **E2E Tests (current file)** — live-server suite
+
+Or run **Tests** from `.vscode/tasks.json` (`poetry run test`).
+
+## Unit vs e2e separation
+
+| Layer | Unit | E2E |
+|-------|------|-----|
+| Location | `tests/` except `tests/e2e/` | `tests/e2e/` only |
+| Marker | unmarked | `@pytest.mark.e2e` |
+| Env | `tests/conftest.py` (dummy values) | `.env` (copy from `.env.template`) |
+| Run command | `test-unit` / `pytest` | `test-e2e` |
+| CI | yes | no |
+
+Default `addopts` run coverage on `src/` and exclude e2e (`-m "not e2e"`).
+
+## Assertions
+
+- Use `pytest.raises(SpecificError)` with the exact exception type.
+- Do not catch or assert against bare `Exception` when a domain error exists.
+
+## Unit and offline tests
+
+- Mock external I/O with fake connectors and lightweight stand-ins (see `tests/inspect/test_actions.py`).
+- Dummy `VIPAT_*` env vars are set in `tests/conftest.py` (autouse fixture; skipped for e2e).
+- Load JSON fixtures from `tests/<app>/fixtures/<version>/` using `pathlib.Path`.
+- Put shared fixtures in `conftest.py` at the appropriate directory level.
+- Use session-scoped fixtures only when setup is expensive and reuse is intentional.
+
+## E2E tests
+
+- Live-server tests live under `tests/e2e/` only.
+- Mark with `@pytest.mark.e2e`; they are excluded from the default suite.
+- E2e entry points (`poetry run test-e2e`, `poetry run test`, VS Code **E2E Tests**) load `.env` and enable the suite automatically. E2e runs use `--no-cov`.
+- Copy `.env.template` to `.env` (gitignored), set connection vars. The e2e conftest loads `.env` automatically when present.
+- Run with `poetry run test-e2e` (no extra env vars on the command line).
+- Namespace all writes with the `E2E-` label prefix and `vipat-e2e` tag.
+- Do not add e2e tests to the default CI/offline run.
+
+## Test data
+
+- All fixture and test data must follow anonymization rules in `AGENTS.md`.
+- Preserve structure and relationships; replace real hostnames, IPs, and customer identifiers with generic placeholders.
+
+## Coverage
+
+- Default runs report coverage on `src/`.
+- Do not disable coverage flags without a clear reason.

.cursor/rules/agent-rules-layout.mdc

@@ -0,0 +1 @@
+../.claude/rules/agent-rules-layout.md

.cursor/rules/python-quality.mdc

@@ -0,0 +1 @@
+../.claude/rules/python-quality.md

.cursor/rules/python-style.mdc

@@ -0,0 +1 @@
+../.claude/rules/python-style.md

.cursor/rules/python-testing.mdc

@@ -0,0 +1 @@
+../.claude/rules/python-testing.md

.env.template

@@ -0,0 +1,13 @@
+# Copy to .env (gitignored) for local development and live-server e2e tests.
+
+VIPAT_ENVIRONMENT=DEV
+VIPAT_VIDEOIPATH_SERVER_ADDRESS=vip-server.example
+VIPAT_VIDEOIPATH_USERNAME=test-user
+VIPAT_VIDEOIPATH_PASSWORD=test-password
+VIPAT_USE_HTTPS=true
+VIPAT_VERIFY_SSL_CERT=false
+VIPAT_LOG_LEVEL=INFO
+VIPAT_ADVANCED_DRIVER_SCHEMA_CHECK=true
+VIPAT_TIMEOUT_HTTP_GET=10
+VIPAT_TIMEOUT_HTTP_PATCH=10
+VIPAT_TIMEOUT_HTTP_POST=10

.github/workflows/ci.yml

@@ -5,7 +5,7 @@ on:
     branches:
       - main
     tags:
-      - 'v*'     # Trigger on version tags like v1.2.3
+      - "v*" # Trigger on version tags like v1.2.3
   pull_request_target:
     branches:
       - main
@@ -20,7 +20,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.11", "3.12", "3.13"]
+        python-version: ["3.11", "3.12", "3.13", "3.14"]
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
@@ -37,7 +37,7 @@ jobs:
         run: poetry install --with dev,test
 
       - name: Run tests
-        run: poetry run pytest
+        run: poetry run test-unit
 
   # 2️⃣ Release Job (Uses Prebuilt Package)
   release:
@@ -52,7 +52,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.13"
+          python-version: "3.14"
 
       - name: Install Poetry
         run: pip install poetry