All Reports
Windows Terminal
C++
microsoft/terminal
Pass Rate
53%
The new Windows Terminal and the original Windows console host
L1100%
L2100%
L365%
L40%
L50%
Strong Task Discovery
Windows Terminal reaches Level 3 with 100% Task Discovery pass rate. Currently becoming autonomous-capable with 29/55 criteria passing (53%). Key areas for improvement include the opportunities listed below.
Strengths
01
Task Discovery (100%)
Includes Backlog Health, Issue Labeling System, Issue Templates.
02
Cyclomatic Complexity
Complexity analysis via C++ Core Check rules in StaticAnalysis.ruleset (includes cognitive complexity checks).
03
Formatter
.clang-format configured with Microsoft style, ColumnLimit 0, specific include ordering rules.
Opportunities
01
Dead Code Detection
Set up dead code detection to keep the codebase lean and maintainable.
02
Test Coverage Thresholds
Establish minimum test coverage thresholds to prevent regression and maintain code quality.
03
Skills
Create .factory/skills to give agents reusable, tested capabilities for common tasks.
All Criteria
Style & Validation4/9 (44%)
—code_modularizationSKIP: C++ project with natural module boundaries via namespaces and compilation units, no additional tooling needed.
✓cyclomatic_complexityComplexity analysis via C++ Core Check rules in StaticAnalysis.ruleset (includes cognitive complexity checks).
✗dead_code_detectionNo dead code detection tools found (no knip, ts-prune, vulture, or C++ equivalent configured).
✗duplicate_code_detectionNo duplicate code detection tools (no jscpd, PMD CPD, or similar configured in CI/pre-commit).
✓formatter.clang-format configured with Microsoft style, ColumnLimit 0, specific include ordering rules.
✗large_file_detectionNo tooling found for detecting large files. No git hooks, LFS configuration, or CI jobs checking file size.
✓lint_configStatic analysis configured via StaticAnalysis.ruleset with C++ Core Check rules (C6001, C6011, etc.).
—n_plus_one_detectionSKIP: Desktop application without database/ORM usage requiring N+1 query detection.
✓naming_consistencyNaming conventions enforced via .editorconfig (indent_style, charset) and clang-format rules.
✗pre_commit_hooksNo pre-commit hooks configured (no Husky, .pre-commit-config.yaml, or git hooks directory).
—strict_typingSKIP: C++ has compile-time type enforcement by default.
✗tech_debt_trackingTODO/FIXME comments exist in code but no automated tracking system (no TODO scanner in CI, no linter rules enforcing issue links).
—type_checkSKIP: C++ is statically typed by default with compile-time type checking.
Build System9/12 (75%)
✗agentic_developmentNo evidence of agent co-authorship in recent 100 commits (no droid, Claude, or co-authored-by agent signatures).
—automated_pr_reviewSKIP: gh CLI available but no evidence of automated PR review bots generating review comments.
✓build_cmd_docBuild commands well-documented in doc/building.md with PowerShell and Cmd examples.
—build_performance_trackingSKIP: gh CLI available but no evidence of build performance tracking or caching metrics analysis in CI.
✗dead_feature_flag_detectionFeature flag infrastructure exists but no tooling to detect stale/unused flags in features.xml.
✓deployment_frequencyMultiple releases per month (30 releases listed in recent history, ~2-4 week cadence between releases).
✓deps_pinnedDependencies pinned via vcpkg.json with explicit version overrides and builtin-baseline commit hash.
✓fast_ci_feedbackCI checks complete in under 10 minutes (spell check: ~2 minutes based on statusCheckRollup timestamps).
✓feature_flag_infrastructureRobust feature flag system using src/features.xml with til::feature, supports branch/branding-based toggles.
—heavy_dependency_detectionSKIP: Native C++ application, not a bundled web app requiring bundle size analysis.
—monorepo_toolingSKIP: Single application repository, not a monorepo with multiple independently deployable packages.
—progressive_rolloutSKIP: Desktop application distributed via Microsoft Store and direct download, not a web service with canary deployments.
✓release_automationAutomated release pipeline (build/pipelines/ob-release.yml) with signing, packaging, and publishing to multiple channels.
✓release_notes_automationAutomated releases with structured release notes (ob-release.yml pipeline, consistent format with bug fixes and contributors).
—rollback_automationSKIP: Desktop application where users control version installation, not an infra-based deployment requiring rollback.
✓single_command_setupSingle command setup documented in doc/building.md: Import-Module + Set-MsBuildDevEnvironment + Invoke-OpenConsoleBuild.
✗unused_dependencies_detectionNo unused dependency detection configured (vcpkg doesn't include built-in detection like npm/cargo).
✓vcs_cli_toolsGitHub CLI (gh) is installed and authenticated (verified with gh auth status).
—version_drift_detectionSKIP: Not a monorepo, no multiple packages requiring version consistency checks.
Testing4/7 (57%)
—flaky_test_detectionSKIP: gh CLI available but no evidence of flaky test detection (no retry config, test stability metrics).
✓integration_tests_existIntegration/feature tests exist (Host.FeatureTests, TerminalApp.LocalTests, WindowsTerminal.UIA.Tests).
✗test_coverage_thresholdsNo test coverage thresholds configured (.runsettings files exist but don't specify coverage requirements).
✗test_isolationNo explicit test isolation configuration (no parallel test execution or randomization settings in .runsettings).
✓test_naming_conventionsConsistent test naming conventions (*UnitTests.vcxproj, *FeatureTests.vcxproj pattern followed across all test projects).
✗test_performance_trackingNo evidence of test performance tracking (no test duration reporting or test analytics platform integration).
✓unit_tests_existExtensive unit tests exist (TextBuffer.Unit.Tests, Host.UnitTests, TerminalApp.UnitTests, etc.).
✓unit_tests_runnableTests runnable via Invoke-OpenConsoleTests PowerShell function and runut.cmd/runft.cmd scripts.
Documentation2/6 (33%)
✗agents_mdNo AGENTS.md file at repository root.
—agents_md_validationSKIP: No AGENTS.md file exists, so validation not applicable.
—api_schema_docsSKIP: Desktop application, not an API service requiring OpenAPI/GraphQL schema documentation.
✗automated_doc_generationNo evidence of automated documentation generation tools (no Doxygen, API doc generators, or doc-gen workflows).
✓documentation_freshnessKey documentation updated recently (CONTRIBUTING.md modified in last 180 days).
✓readmeComprehensive README.md with installation, building, running, debugging, and contribution instructions.
✗service_flow_documentedNo architecture diagrams found (no .mermaid, .puml files, or architecture/diagrams directories).
✗skillsNo skills directory found (.factory/skills/, .skills/, .claude/skills/).
Dev Environment0/1 (0%)
—database_schemaSKIP: No database usage (terminal emulator with in-memory state, no Prisma/TypeORM/SQL schemas).
✗devcontainerNo .devcontainer/devcontainer.json configuration found.
—devcontainer_runnableSKIP: No devcontainer configuration exists to test.
—env_templateSKIP: Desktop application that doesn't require environment variable configuration (no .env usage in codebase).
—local_services_setupSKIP: No external service dependencies (no database, Redis, etc.) requiring docker-compose or local setup.
Debugging & Observability3/8 (38%)
✗alerting_configuredNo alerting system configured (no PagerDuty, OpsGenie, or custom alerting rules found).
—circuit_breakersSKIP: Desktop application without external service dependencies requiring circuit breaker patterns.
—code_quality_metricsSKIP: gh CLI authenticated but insufficient permissions (403) to verify code scanning/quality metrics via API.
✗deployment_observabilityNo deployment observability references (no monitoring dashboard links in docs, no deploy annotations).
✓distributed_tracingDistributed tracing via ETW TraceLogging with activity IDs and correlation (TRACELOGGING_DECLARE_PROVIDER infrastructure).
✗error_tracking_contextualizedNo contextualized error tracking (Sentry, Bugsnag, Rollbar not found in dependencies or configuration).
—health_checksSKIP: Desktop application not deployed as a service, health check endpoints not applicable.
✓metrics_collectionTelemetry/metrics collection via ProjectTelemetry.h (TraceLoggingOptionMicrosoftTelemetry, TelemetryPrivacyDataTag macros).
✗profiling_instrumentationNo profiling instrumentation found (no APM tools, continuous profiling, or profiling framework configured).
✗runbooks_documentedNo runbooks or incident response procedures documented or referenced in README/docs.
✓structured_loggingStructured logging via Windows ETW TraceLogging (tracing.hpp files, ProjectTelemetry.h infrastructure).
Security3/6 (50%)
—automated_security_reviewSKIP: gh CLI authenticated but API returns 403 for code-scanning/alerts (insufficient permissions to verify).
✓branch_protectionBranch protection via GitHub rulesets (2 active rulesets found: microsoft-production-ruleset and check-spelling).
✗codeownersNo CODEOWNERS file found in root or .github/ directory.
—dast_scanningSKIP: Desktop application, not a web service requiring DAST scanning (OWASP ZAP, Burp Suite).
✗dependency_update_automationNo Dependabot or Renovate configuration found (.github/dependabot.yml, renovate.json missing).
✓gitignore_comprehensiveComprehensive .gitignore excludes .env files, node_modules, build artifacts, IDE configs (.idea, .vscode), and OS files (.DS_Store).
✗log_scrubbingNo log scrubbing/sanitization configured (TraceLogging exists but no redaction or filtering mechanisms found).
—pii_handlingSKIP: Terminal emulator that doesn't collect or process PII (no user data storage or transmission).
—privacy_complianceSKIP: Desktop application without end-user data collection (terminal emulator, no PII processing requirements).
—secret_scanningSKIP: gh CLI authenticated but API returns 404 for secret-scanning/alerts endpoint, cannot verify enablement.
✓secrets_managementNo hardcoded secrets found, .env properly gitignored, uses Azure DevOps pipeline secrets for signing credentials.
Task Discovery4/4 (100%)
✓backlog_health100% of open issues (50 sampled) have descriptive titles >10 chars AND at least one label.
✓issue_labeling_systemConsistent labeling system (Issue-Bug, Needs-Triage, Product-Terminal, Area-* for components).
✓issue_templatesIssue templates configured in .github/ISSUE_TEMPLATE/ (Bug_Report.yml, Feature_Request.yml, config.yml).
✓pr_templatesPR template exists at .github/PULL_REQUEST_TEMPLATE.md with sections for summary, references, validation, and checklist.
Product & Analytics0/2 (0%)
✗error_to_insight_pipelineNo error-to-insight pipeline (no Sentry-GitHub integration, no automated issue creation from errors).
✗product_analytics_instrumentationNo product analytics instrumentation (Mixpanel, Amplitude, PostHog, Heap, GA4 not found in dependencies).