Structure prediction for protein, nucleic-acid, and small-molecule complexes with the Chai-1 foundation model (Chai Discovery 2024, github.com/chaidiscovery/chai-lab). Reach for this skill to predict an antibody-antigen or protein-ligand complex from a single FASTA, to re-fold designed binders as an AlphaFold-multimer alternative, or to drive co-folding from Python for batched campaigns on a GPU.

Working pattern for Grida Gateway (GG) — Grida's first-party, metered, no-BYOK AI surface: the scoped-token mint, the OpenAI-compatible + native gateway endpoints, the `gg` client provider, and the desktop wiring that spends org credit without a user key. Anchor for the `GRIDA-GG: <surface>` grep marker; its security half is `GRIDA-SEC-006`. Use when adding or touching any GG file, the `gg` provider kind, the `gg:ai` token audience, or deciding whether code belongs to the gateway surface. Companions: `security` (the GRIDA-SEC-006 half), `ee-billing` (the ledger it spends), `agent-system` and `desktop` (the two hosts).

Surface workflow for billing in Grida — Stripe (subscriptions) + Metronome (AI credit ledger). The stable contracts: `grida_billing.*` is not REST-exposed (views/RPCs only), `fn_billing_apply_*` are the single mutation points, webhook receivers are `GRIDA-SEC-001`, BYOK is the `GRIDA-SEC-003` carve-out. Use when touching `editor/lib/billing/`, `editor/scripts/billing/`, the `grida_billing` schema, the webhook receivers, or the entitlement gate. Companion to `ee`.

Decide which family a doc belongs to before drafting — SDK/developer, user/product, or working-group (WG) — since family sets the audience, home, and tone. Use when creating, moving, or restructuring docs, when unsure which directory a doc belongs in, or when a request says "document this" / "write docs for X" without naming the kind. Routes to the specialized skill for each family.

Doctrine for drafting and keeping working-group docs under `docs/wg/**` — RFC/RFD specs and findings/research/glossary. A WG doc is a language-agnostic, code-agnostic study of a domain: it argues *why* and defines *what*, never *how in our code*. Use when writing or editing anything under `docs/wg/`, an RFC/RFD, a spec, a design note, a glossary, or research findings — including "write up the design", "document the spec", or "capture what we learned". Not for plans/TODOs (untracked `*.plan.md`), user docs, or SDK API refs — use `docs` to route those.

RT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with competitive accuracy and supports distillation and quantization for deployment optimization. Use when training, evaluating, distilling, quantizing, exporting, or running inference for a TAO RT-DETR model. Trigger phrases include "train RT-DETR", "real-time DETR", "low-latency object detection", "RT-DETR distillation / quantization".

Metric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for retrieval-based matching (e.g., retail product recognition) using triplet / contrastive losses. Use when training, evaluating, exporting, or running inference for a TAO metric-learning recognition model. Trigger phrases include "train metric learning", "ml-recog", "retrieval embeddings", "triplet loss recognition", "fine-grained matching".

Start, query, and stop a network-specific TAO inference microservice ({network_arch}-inference-microservice) by delegating container execution to the appropriate platform skill. Handles container image resolution, job-payload JSON construction, and the service registry. Use when the user wants to run inference on a TAO model checkpoint using a microservice container, deploy a TAO inference endpoint, or stop a running inference container.

Run `tao-daft convert` to convert NVIDIA TAO DAFT datasets between supported formats. Do not use for non-DAFT data. Use when the user asks to convert a DAFT dataset, change DAFT format, change a TAO dataset format, or run `tao-daft convert`.

Use this skill when choosing or configuring NeMo Relay 0.6 or 0.7 observability through the built-in plugin, subscribers, or exporters, including raw ATOF events, ATIF trajectories, OpenTelemetry, OpenInference, or custom event handling.

Use this skill when building or packaging reusable NeMo Relay runtime behavior as an embedded configuration component or a manifest-backed `rust_dynamic` native or `worker` gRPC plugin, with deterministic validation and rollback-safe registration.

Use this skill when choosing or running NeMo Relay installation for the CLI, Python, Node.js, Rust, OpenClaw, Hermes, or maintained framework integrations before runtime configuration or quick-start setup.

Profile a DeepStream pipeline with Nsight Systems and derive its configs from the measurement. Use when the user asks for an efficient, performant, or profiled pipeline — or to benchmark, tune, or measure FPS.

Pre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment, scaling analysis, or refactor planning. Inspect the user's source code, look up NumPy usage, cross-reference the cuPyNumeric API support manifest, and distinguish distributed-scaling-friendly patterns from blockers such as unsupported APIs, scalar synchronization, host round-trips, Python/object-heavy control flow, shape/data-dependent branching, and in-place mutation hazards. Produce a verdict of READY, LIGHT REFACTOR, SIGNIFICANT REFACTOR, or NOT RECOMMENDED, with concrete refactor pointers.

LP, MILP, and QP (beta) with cuOpt — Python, C, and CLI. Use when the user is solving LP, MILP, or QP with any cuOpt interface.

Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.

Add `flutter_localizations` and `intl` dependencies, enable "generate true" in `pubspec.yaml`, and create an `l10n.yaml` configuration file. Use when initializing localization support for a new Flutter project.

Guide agents to use `package:ffigen` to automatically generate FFI bindings instead of writing them manually. Use this skill when a task involves writing new FFI bindings, extending C/Objective-C/Swift integrations, or replacing hand-crafted `dart:ffi` setups.

Define and generate mock objects for external dependencies using `package:mockito` and `build_runner`. Use when unit testing classes that depend on complex external services like APIs or databases.

Search, read, and extract attachments from Apple Mail's local storage. Query emails by sender, recipient, subject, body, date, mailbox, and flags. Read raw RFC822 messages and extract file attachments.