Use when designing, adding, reviewing, or debugging Meteor E2E tests in tools/e2e-tests, or maintaining their app/skeleton coverage report. Covers fixture selection, meaningful assertions, lifecycle helpers, isolation, CI grouping, and coverage documentation. Use testing for shared test-value and change-scope rules and choosing another test layer.

Create a new worktrunk worktree (optionally in another repo) and switch this session's working directory into it. Use when launching a session that should work in its own worktree.

When the user wants to implement, optimize, or audit favicon and app icons. Also use when the user mentions "favicon," "app icon," "browser icon," "touch icon," "PWA icon," "favicon sizes," "apple-touch-icon," "favicon.ico," "site icon," or "tab icon." For visual system, use brand-visual-generator.

Build managed ingestion pipelines into Databricks using Lakeflow Connect. Use when ingesting from SaaS apps (Salesforce, Workday Reports, ServiceNow, Google Analytics 4, HubSpot, Confluence) or databases (SQL Server cloud and on-prem; PostgreSQL/MySQL CDC in PuPr) into Unity Catalog with serverless pipelines.

Apache Iceberg tables on Databricks — Managed Iceberg tables, External Iceberg Reads (fka Uniform), Compatibility Mode, Iceberg REST Catalog (IRC), Iceberg v3, Snowflake interop, PyIceberg, OSS Spark, external engine access and credential vending. Use when creating Iceberg tables, enabling External Iceberg Reads (uniform) on Delta tables (including Streaming Tables and Materialized Views via compatibility mode), configuring external engines to read Databricks tables via Unity Catalog IRC, integrating with Snowflake catalog to read Foreign Iceberg tables

Databricks AI Search (formerly Vector Search) endpoints and indexes for RAG and semantic search. Use when asked to: create Standard or Storage-Optimized endpoints and Delta Sync or Direct Access indexes with AISearchClient; query with ANN, hybrid, or full-text search and filters; sync, monitor, or migrate indexes; set a target QPS, triage 429s, or load-test an endpoint; wire an index into an agent as a retriever tool. NOT for: no-code document Q&A (databricks-genie-agents); serving-endpoint CRUD (databricks-model-serving); custom ResponsesAgent authoring and deployment (databricks-ml-training).

Databricks AI Runtime, the `databricks air` CLI commands for submitting and managing GPU training workloads on Databricks serverless compute. Use for: writing and submitting `databricks air` workload YAML, choosing an accelerator type and Python environment, passing hyperparameters and secrets, configuring retries, checking run status, listing/cancelling runs, streaming a run's logs and watching its progress, debugging failed runs, GPU pools, multi-container workloads, custom Docker image setup, and converting a workload to a Databricks Asset Bundle.

Apache Iceberg tables on Databricks — Managed Iceberg tables, External Iceberg Reads (fka Uniform), Compatibility Mode, Iceberg REST Catalog (IRC), Iceberg v3, Snowflake interop, PyIceberg, OSS Spark, external engine access and credential vending. Use when creating Iceberg tables, enabling External Iceberg Reads (uniform) on Delta tables (including Streaming Tables and Materialized Views via compatibility mode), configuring external engines to read Databricks tables via Unity Catalog IRC, integrating with Snowflake catalog to read Foreign Iceberg tables

Use when writing or revising discount rules: discount matrix, approver thresholds, exception flows and the deal framework sales works under. Builds the matrix with discount_matrix_builder.py, routes exceptions with exception_router.py and checks the policy with policy_linter.py.

Use when a founder wants to document or design a whole company as version-controlled markdown (company as code, a company wiki agents can read, an OKF bundle). Runs a 12-phase interview, scaffolds with scaffold_bundle.py and checks it with okf_linter.py.

Use when a founder wants to document or design a whole company as version-controlled markdown (company as code, a company wiki agents can read, an OKF bundle). Runs a 12-phase interview, scaffolds with scaffold_bundle.py and checks it with okf_linter.py.

Use when authoring or auditing a skill in this ecosystem, before release. Validates structure with skill_validator.py, runs its scripts with script_tester.py and grades quality and security with quality_scorer.py and security_scorer.py.

Use for CAPA work in a medical device QMS: root cause analysis (5-Why, fishbone), corrective action plans, effectiveness checks or CAPA metrics. Tracks CAPAs with capa_tracker.py and analyzes causes with root_cause_analyzer.py.

Grade finished work honestly on two axes with a devil's-advocate pass against inflation. Use whenever the user asks how well a task went, to rate the work or for an honest self-review.

Use when converting a markdown code review with diff blocks and severity callouts into a two-column HTML review. Parses diffs with diff_parser.py and annotations with annotation_extractor.py; requires a named reviewer.

Build a scraper or extractor for sites, PDFs or paginated APIs and validate the fields (Firecrawl or local Python). Load whenever the user wants data pulled from pages, files or an API, or a scraper broke.

Use for ISO/IEC 42001 AI management system work: gap analysis against Clauses 4-10, the AI risk register and Annex A controls, or the internal audit plan. Runs aims_gap_analyzer.py, ai_risk_register_builder.py and aims_audit_scheduler.py.

Use before adding memory to an agent or when a memory store keeps growing (long context vs RAG vs graph vs agentic memory, what to expire). Prices the design with memory_cost_profiler.py and refuses one without a forgetting policy via forgetting_policy_linter.py.

Use before shipping anything that needs human sign-off (a plan, spec, report, page, migration or irreversible action). Builds a review page with review_page_builder.py and runs human_gate.py, which refuses to close with an open blocker or no named reviewer.

Test-first development and test generation for Jest, Pytest, JUnit, Vitest or Mocha: red-green-refactor, fixtures and mocks, coverage gaps. Use whenever the user asks to write tests (before or after the code), mock a dependency or raise coverage.