superbrowser-sdk 2.0.2__tar.gz → 2.1.1__tar.gz

superbrowser_sdk-2.1.1/.github/workflows/benchmark.yml

@@ -0,0 +1,49 @@
+name: Benchmark
+
+on:
+  workflow_dispatch:
+    inputs:
+      iterations:
+        description: "Iterations per metric"
+        required: false
+        default: "5"
+
+jobs:
+  benchmark:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install package with patchright
+        run: pip install -e ".[dev,patchright]"
+
+      - name: Install browser
+        run: python -m patchright install chromium
+
+      - name: Install OS deps for Chromium
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y libnss3 libnspr4 libatk1.0-0 libatk-bridge2.0-0 \
+            libcups2 libdrm2 libdbus-1-3 libxkbcommon0 libatspi2.0-0 libxcomposite1 \
+            libxdamage1 libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 libcairo2 libasound2
+
+      - name: Run benchmark
+        run: |
+          python scripts/browser_benchmark.py \
+            --fixtures benchmarks/fixtures \
+            --out-dir benchmarks/results \
+            --iterations ${{ github.event.inputs.iterations }} \
+            --warmup 1
+
+      - name: Upload benchmark results
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: benchmark-results
+          path: benchmarks/results/
+          retention-days: 90

superbrowser_sdk-2.1.1/.github/workflows/e2e.yml

@@ -0,0 +1,62 @@
+name: E2E Real-Browser Tests
+
+on:
+  workflow_dispatch:
+    inputs:
+      live:
+        description: "Run live navigation tests (requires network)"
+        required: false
+        default: "false"
+        type: choice
+        options:
+          - "false"
+          - "true"
+
+jobs:
+  e2e:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install package with patchright
+        run: pip install -e ".[dev,patchright]"
+
+      - name: Install browser
+        run: python -m patchright install chromium
+
+      - name: Install OS deps for Chromium
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y libnss3 libnspr4 libatk1.0-0 libatk-bridge2.0-0 \
+            libcups2 libdrm2 libdbus-1-3 libxkbcommon0 libatspi2.0-0 libxcomposite1 \
+            libxdamage1 libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 libcairo2 libasound2
+
+      - name: Run E2E tests
+        env:
+          SB_E2E: "1"
+          SB_E2E_LIVE: ${{ github.event.inputs.live == 'true' && '1' || '' }}
+          SB_HEADLESS: "1"
+        run: |
+          python -m pytest tests/e2e/ -v --tb=short
+
+      - name: Upload E2E reports
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: e2e-reports
+          path: tests/e2e/artifacts/
+          retention-days: 30
+
+      - name: Upload failure screenshots
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: e2e-screenshots
+          path: tests/e2e/artifacts/*-failure.png
+          if-no-files-found: ignore
+          retention-days: 14

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/.github/workflows/publish.yml

@@ -24,9 +24,12 @@ jobs:
       - name: Build package
         run: python -m build
 
-      - name: Verify package
+      - name: Verify package (twine)
         run: twine check dist/*
 
+      - name: Verify release artifact (guard)
+        run: python scripts/verify_release_artifact.py dist/superbrowser_sdk-*.whl README.md docs/
+
       - name: Publish to PyPI
         uses: pypa/gh-action-pypi-publish@release/v1
         # Uses trusted publisher (no API token needed)

superbrowser_sdk-2.1.1/.github/workflows/smoke.yml

@@ -0,0 +1,74 @@
+name: Post-Release Smoke
+
+on:
+  schedule:
+    # Run daily at 08:00 UTC to catch PyPI-side breakages early
+    - cron: "0 8 * * *"
+  workflow_dispatch:
+    inputs:
+      version:
+        description: "Specific version to smoke test (default: latest)"
+        required: false
+        default: ""
+
+jobs:
+  smoke:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    permissions:
+      contents: read
+      issues: write
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Run smoke test
+        run: |
+          ARGS=""
+          if [ -n "${{ github.event.inputs.version }}" ]; then
+            ARGS="--version ${{ github.event.inputs.version }}"
+          fi
+          python scripts/smoke_published.py \
+            --out smoke-results/smoke-report.json \
+            --md smoke-results/smoke-report.md \
+            $ARGS
+
+      - name: Upload smoke report
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: smoke-report
+          path: smoke-results/
+          retention-days: 30
+
+      - name: Open issue on failure
+        if: failure() && github.event_name == 'schedule'
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const fs = require('fs');
+            let report;
+            try {
+              report = JSON.parse(fs.readFileSync('smoke-results/smoke-report.json', 'utf-8'));
+            } catch (e) {
+              report = { summary: { failed: 'unknown' } };
+            }
+            const body = [
+              `## Automated Post-Release Smoke Failure`,
+              ``,
+              `**Trigger:** ${{ github.event_name }}`,
+              `**Run:** ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}`,
+              `**Failed checks:** ${report.summary?.failed ?? 'unknown'}`,
+              ``,
+              `See the [smoke report artifact](../actions/runs/${{ github.run_id }}) for details.`,
+            ].join('\n');
+            await github.rest.issues.create({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              title: `🚨 Post-release smoke failure (${new Date().toISOString().split('T')[0]})`,
+              body,
+              labels: ['bug', 'release'],
+            });

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/.gitignore

@@ -61,3 +61,16 @@ NUL
 sessions/
 sources/
 events.jsonl
+
+# E2E test artifacts
+tests/e2e/artifacts/
+
+# Benchmark results
+benchmarks/results/
+
+# Smoke test results
+smoke-results/
+
+# E2E trend artifacts
+tests/e2e/artifacts/e2e-history.json
+tests/e2e/artifacts/e2e-trend.md

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/CHANGELOG.md

@@ -7,6 +7,93 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [2.1.1] — 2026-06-16
+
+### Fixed — Smoke workflow install spec format
+
+- **Smoke install specs were malformed** (`dist==version[extra]` instead
+  of `dist[extra]==version`), causing the post-release smoke workflow
+  to fail for any version-targeted run.
+- Affects `[all]`, `[patchright]`, and `[playwright]` install checks in
+  `scripts/smoke_published.py`.
+- Found by the v2.1.0 post-release smoke run — exactly the kind of
+  packaging bug the workflow was designed to catch.
+- Added `TestInstallSpecFormat` regression test verifying PEP 508
+  extras-before-version ordering.
+
+### Fixed — Benchmark deprecated API usage
+
+- `scripts/browser_benchmark.py` used the deprecated `raw_page` attribute,
+  producing `DeprecationWarning` on every run.
+- Replaced `page.raw_page.query_selector_all` → `page.backend_page`
+- Replaced `page.raw_page.set_viewport_size` → `page.backend_page`
+
+### Version metadata
+
+- `pyproject.toml`, `__init__.py`, doc headers, test assertions bumped
+  to `2.1.1`.
+
+## [2.1.0] — 2026-06-16
+
+### Overview
+
+**v2.1.0 — Operational Maturity**
+
+v2.1 makes Super Browser easier to operate, validate, benchmark, and
+monitor after release. No new stealth features, no API redesigns —
+this release adds the operational infrastructure needed to trust the
+E2E signal, catch packaging regressions before users do, and track
+performance over time.
+
+### Added — Benchmarking (Track A)
+
+- Offline real-browser benchmark harness (`scripts/browser_benchmark.py`)
+  measuring 9 metrics: browser launch, new page, navigation (simple/form/
+  DOM-heavy), DOM query, click+fill, screenshot, and memory delta.
+- Local fixture-backed pages only — no network dependency.
+- JSON schema v1 output with mean/median/min/max/stdev per metric.
+- Markdown report for human-readable consumption.
+- Configurable iterations, warmup, headless mode, backend selection.
+- `timeout_s` enforcement via `asyncio.wait_for` on all operations.
+- Fixture HTTP server cleanup in `try/finally`.
+- Manual `benchmark.yml` workflow (`workflow_dispatch` only, 90-day artifacts).
+
+### Added — Release Observability (Track B)
+
+- Post-release PyPI smoke workflow (`scripts/smoke_published.py`).
+- Fresh-venv install validation: `[all]`, `[patchright]`, `[playwright]`,
+  `import super_browser`, `superbrowser version`, `superbrowser info`.
+- Version match check when `--version` specified.
+- JSON + Markdown smoke report output.
+- Daily scheduled run at 08:00 UTC + manual `workflow_dispatch`.
+- Automatic GitHub issue creation on scheduled failures.
+- Report always written, even on early failures (venv creation, install).
+- Explicit `issues: write` permission for auto-issue behavior.
+
+### Added — E2E Reporting (Track C)
+
+- **Schema v3**: Formalized E2E report contract with stdlib validator
+  (`scripts/validate_e2e_report.py`). Required: `schema_version`,
+  `timestamp_utc`, `environment`, `config`, `summary`, `tests`, `artifacts`.
+  Per-test entries: `name`, `status`, `duration_s`, `file`, `error`,
+  `screenshot`.
+- **Failure metadata**: `error` field populated from `call.excinfo` /
+  `report.longrepr`. `screenshot` field populated on failure, attached by
+  `nodeid` match. Both nullable for passed/skipped tests.
+- **Historical trend artifacts** (`scripts/e2e_trend.py`): Archives
+  validated reports into `e2e-history.json` (max 30 runs, dedup by
+  timestamp). Renders `e2e-trend.md` with pass-rate, duration, budget, and
+  failure trends.
+- **Lifecycle integration**: Report validation, failure metadata capture,
+  and trend generation all wired into `pytest_sessionfinish`.
+
+### Operational Safety
+
+- No benchmark performance thresholds (measurement only).
+- No E2E regression gates (observational trends).
+- No live tests enabled by default.
+- Default CI unaffected by any new workflow.
+
 ## [2.0.2] — 2026-06-15
 
 ### Fixed — CLI entry point (#148)

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: superbrowser-sdk
-Version: 2.0.2
+Version: 2.1.1
 Summary: Anti-detection agent browser SDK — stealth, budget governance, security guardrails, and structured error recovery behind one async facade
 Project-URL: Homepage, https://github.com/Octo-Lex/Super-Browser
 Project-URL: Documentation, https://github.com/Octo-Lex/Super-Browser#readme

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/api-reference.md

@@ -1,6 +1,6 @@
 # API Reference
 
-> **Super Browser** v2.0.0a1 — Complete public API documentation.
+> **Super Browser** v2.1.1 — Complete public API documentation.
 
 This reference covers every public class, method, and function in Super Browser. Classes are grouped by subsystem.
 

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/architecture.md

@@ -1,6 +1,6 @@
 # Architecture
 
-> **Super Browser** v2.0.0a1 — System architecture, data flow, and extension points.
+> **Super Browser** v2.1.1 — System architecture, data flow, and extension points.
 
 ---
 

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/fingerprint-scoring.md

@@ -172,13 +172,13 @@ The `stealth-check` CLI command provides quick fingerprint assessment:
 
 ```bash
 # Offline check (default) — no browser needed
-super-browser stealth-check
+superbrowser stealth-check
 
 # With options
-super-browser stealth-check --format html --threshold 80
+superbrowser stealth-check --format html --threshold 80
 
 # Online check — requires live browser
-super-browser stealth-check --online
+superbrowser stealth-check --online
 ```
 
 ### Options

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/fingerprint-validation.md

@@ -30,13 +30,13 @@ The fingerprint validation suite provides 8 cross-surface consistency checks tha
 
 ```bash
 # Capture baseline for current profile
-super-browser stealth-validate --capture-baseline
+superbrowser stealth-validate --capture-baseline
 
 # Run validation against baseline
-super-browser stealth-validate --profile windows-chrome-stable --seed my-session
+superbrowser stealth-validate --profile windows-chrome-stable --seed my-session
 
 # CI mode: exit code 1 on any regression
-super-browser stealth-validate --ci
+superbrowser stealth-validate --ci
 ```
 
 ## Programmatic Usage
@@ -62,8 +62,8 @@ for check in report.checks:
 
 The `--ci` flag compares current results against a saved baseline:
 
-1. First run: `super-browser stealth-validate --capture-baseline` saves baseline JSON
-2. CI run: `super-browser stealth-validate --ci` compares and exits 1 on regression
+1. First run: `superbrowser stealth-validate --capture-baseline` saves baseline JSON
+2. CI run: `superbrowser stealth-validate --ci` compares and exits 1 on regression
 3. Baselines stored in `~/.config/super-browser/baselines/{profile_id}.json`
 
 A check is a **regression** if it passed in the baseline but fails now. Checks that were already failing are not flagged as regressions.

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/memory.md

@@ -224,16 +224,16 @@ Memory can be managed from the command line:
 
 ```bash
 # List all domains with stored memory
-super-browser memory list
+superbrowser memory list
 
 # Show memory for a specific domain
-super-browser memory show shop.example.com
+superbrowser memory show shop.example.com
 
 # Clear memory for a domain
-super-browser memory clear shop.example.com
+superbrowser memory clear shop.example.com
 
 # Prune expired entries
-super-browser memory prune
+superbrowser memory prune
 ```
 
 ---

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/docs/quickstart.md

@@ -400,3 +400,77 @@ Ensure you have a Chromium-based browser installed:
 pip install superbrowser-sdk[patchright]
 python -m patchright install chromium
 ```
+
+## E2E Tests (Real-Browser)
+
+Super Browser ships with an opt-in E2E test suite that exercises real
+browser interactions against local fixture pages. These tests are **not**
+run in default CI — you enable them explicitly.
+
+### Prerequisites
+
+```bash
+pip install superbrowser-sdk[patchright]
+python -m patchright install chromium
+```
+
+### Running locally
+
+```bash
+# Run all local E2E tests (headless)
+SB_E2E=1 pytest tests/e2e/
+
+# Run with visible browser (debugging)
+SB_E2E=1 SB_HEADLESS=0 pytest tests/e2e/
+
+# Include live navigation tests (requires network)
+SB_E2E=1 SB_E2E_LIVE=1 pytest tests/e2e/
+```
+
+### Environment variables
+
+| Variable | Default | Description |
+|:---------|:--------|:-------------|
+| `SB_E2E` | unset | Enables E2E tests. If unset, all tests skip. |
+| `SB_E2E_LIVE` | unset | Enables live navigation tests (external network). |
+| `SB_HEADLESS` | `1` | Set to `0` for headed browser mode (debugging). |
+| `SB_BACKEND` | `patchright` | Browser backend. |
+| `SB_E2E_BUDGET_S` | `120` | Suite-level time budget in seconds. |
+| `SB_E2E_REPORT_DIR` | `tests/e2e/artifacts/` | Report output directory. |
+
+### Reports
+
+After a run, JSON and Markdown reports are written to
+`tests/e2e/artifacts/`:
+
+- `e2e-report.json` — structured results (schema v3, validated)
+- `e2e-report.md` — human-readable summary
+- `{test_name}-failure.png` — screenshots on failure (best-effort)
+
+The JSON report follows schema v3. You can validate any report file:
+
+```bash
+python scripts/validate_e2e_report.py tests/e2e/artifacts/e2e-report.json
+```
+
+After each E2E run, the trend tracker archives the report summary into
+`e2e-history.json` and renders `e2e-trend.md` showing pass-rate,
+duration, and budget trends across recent runs. You can also run it
+manually:
+
+```bash
+python scripts/e2e_trend.py \
+  --reports tests/e2e/artifacts/e2e-report.json \
+  --history tests/e2e/artifacts/e2e-history.json \
+  --markdown tests/e2e/artifacts/e2e-trend.md \
+  --max-runs 30
+```
+
+### CI integration
+
+E2E tests can be triggered manually via GitHub Actions:
+
+1. Go to **Actions** → **E2E Real-Browser Tests**
+2. Click **Run workflow**
+3. Optionally enable live tests
+4. Reports are uploaded as workflow artifacts (30-day retention)

{superbrowser_sdk-2.0.2 → superbrowser_sdk-2.1.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "superbrowser-sdk"
-version = "2.0.2"
+version = "2.1.1"
 description = "Anti-detection agent browser SDK — stealth, budget governance, security guardrails, and structured error recovery behind one async facade"
 readme = "README.md"
 license = { text = "Apache-2.0" }