Loading...
Loading...
Found 3,814 Skills
End-to-end feature addition: takes a feature request in plain English and incrementally writes ***plain specs (concepts, implementation reqs, functional specs, acceptance tests) one functionality at a time, asking, authoring, and reviewing per functionality. Use when the user wants to add a new feature to an existing project.
Bright Security integration. Manage data, records, and automate workflows. Use when the user wants to interact with Bright Security data.
This skill should be used when the user asks to forecast aggregate sentiment and opinion dynamics over time—sentiment indices from text streams; temporal rollups; leading/lagging KPI links; time-series and sequence models (ARIMA, Prophet, state-space, ML); nowcasting; spikes, bots, and bias; walk-forward backtests; intervals and scenarios; volume/velocity/topic features; BI or brand dashboards. Triggers: sentiment forecasting, forecast sentiment, sentiment index, opinion trend forecast, social sentiment time series, brand sentiment trajectory, nowcast sentiment, sentiment leading indicator, aggregate polarity forecast, sentiment backtest, walk-forward sentiment, sentiment spike prediction. Not for per-text labeling (sentiment-analysis-engineer), demand forecasting without sentiment (predictive-logistics-developer, data-scientist), trade advice (methodology only), marketing copy (content-creator), macro without text sentiment (financial-analyst partial).
This skill should be used when the user asks to "create a workflow", "create a getlark test", "add an end-to-end test", "author a larkci workflow", or runs `/getlark:create-workflow`. Converts a natural-language test description (target + ordered steps; target may be a URL, API endpoint, CLI binary, script, or any other software surface) into a `getlark workflows create` invocation with an auto-generated name. Prefer `manage` when the user wants to update or archive an existing workflow, and `invoke-workflow` when they want to run one — this skill only *creates* new workflows.
Add a new cuTile GPU kernel operator to TileGym. Covers dispatch registration in ops.py, cuTile backend implementation, __init__.py exports, test creation, and benchmark in tests/benchmark. Use when adding, creating, or implementing a new cuTile operator/kernel in TileGym, or when asking how to register a new cuTile op.
Head-to-head comparison of coding agents (Claude Code, Aider, Codex, etc.) on custom tasks with pass rate, cost, time, and consistency metrics
Pre-Production validation — build a production-quality end-to-end build to confirm the full game loop is achievable before committing to Production. Run after GDDs, architecture, and UX specs are complete. Produces a PROCEED/PIVOT/KILL verdict that gates the Pre-Production → Production transition.
Use for simulator lifecycle, app install/launch, live viewing, UI inspection, touch/keyboard automation, screenshots, recordings, logs, pasteboard, hardware controls, and repeatable simulator flows.
The meta skill. Turn any raw feature into a properly-skilled, tested, resolvable unit of agent capability. Cross-modal eval is the recommended Phase 3 quality gate: 3 frontier models from different providers critique the output, you iterate to quality, THEN write tests that lock in the proven-good behavior.
Go programming patterns and idioms
Use when the user wants a non-interactive Grill Me run written as one collapsible markdown document, asks for "Grill Me Auto", "auto grill", "all grill questions at once", or invokes /grill-me-auto, /Grill Me Auto, or /grill me auto.
Owns the smoke test contract for an ML experiment: a small, diagnostic-by-construction pytest that fits the experiment's learner on a portion of the real `data/` source and predicts on a *disjoint* portion that deliberately carries **no pre-history buffer**. The assertion is structural — the number of predictions must equal the number of rows in the predict grid. A pipeline that loads-then-features-then-splits will silently drop the cold-start rows of the predict slice and the test will fail with a row-count mismatch; a pipeline that marks X early and references upstream history nodes from feature steps will pass trivially. The smoke test is the executable proof of the X-marker placement rule from `build-ml-pipeline`. TRIGGER when: `test-ml-pipeline` has dispatched here to write the smoke test for an approved experiment; `pytest tests/smoke/` is failing on row count; the user asks "why is the smoke test failing?"; a pipeline edit in `build-ml-pipeline` needs an executable proof; an experiment script changes the pipeline shape and the matching smoke test needs revisiting. SKIP when: the design note does not exist or is not yet approved (route to `iterate-ml-experiment`); the user is asking about a regression test or schema invariant (route to `regression-test-ml-pipeline` / `distribution-test-ml-pipeline` once those exist); the question is the *interpretation* of CV metrics, not predict-time correctness (route to `evaluate-ml-pipeline`). HOW TO USE: read the matching experiment's `journal/NN_*.md` and `experiments/NN_*.py` first to understand the pipeline's source binding (what env-dict keys does `build_learner` expect?). Then construct two env-dicts from the **real `data/` source** — a train env and a predict env — such that the predict env carries *only the rows we want predictions for* and *no pre-history buffer*. The hard assertion is that the prediction count matches the predict-env row count exactly. The soft assertion is that the smoke set's MAE is within `3 × CV_mean` (or the task-appropriate analogue). **Do not write the design note or run CV — that's other skills' job.**