Loading...
Loading...
Found 4,647 Skills
Detect AI-generated code patterns ("slop") in PHP/Laravel and TypeScript/React source — comment narration, generic naming, premature interfaces, defensive overdose, mock-everything tests, and the absence of human "scars". Use when reviewing AI-assisted PRs, auditing code for taste/quality (not metrics — that's technical-debt), or hardening a code-review checklist. Triggers on "review for AI slop", "find AI patterns", "check code feels human", "audit code-quality taste".
Diagnostic guide for active Prometheus cardinality problems — slow queries, OOMing Prometheus, high Grafana Cloud Active Series or DPM bills, "too many samples" ingest errors, series churn, or rapid memory growth. Walks through tsdb status endpoints, per-metric and per-label drill-downs, common-culprit galleries, and remediation paths. Use when the user is *currently experiencing* a cardinality fire. For preventing cardinality issues at the source, route to prometheus-label-strategy. For post-ingest aggregation, route to adaptive-metrics. For DPM-specific analysis, route to dpm-finder.
Comprehensive Rust code review across four lenses — source code (ownership, borrowing, lifetimes, errors, trait design, unsafe, common mistakes), tests (unit, integration, async testing, mocking, property-based), tokio async (task management, sync primitives, channels), and FFI (extern blocks,
Core Power BI data modeling, source connectivity, and platform fundamentals. PROACTIVELY activate for: (1) Power BI data modeling and star-schema design, (2) relationships (active/inactive, bidirectional, USERELATIONSHIP), (3) data-source selection (DirectQuery vs Import vs Direct Lake vs composite), (4) incremental refresh setup, (5) gateway configuration (on-prem and VNet gateways), (6) streaming datasets and push-data scenarios, (7) Dataflow Gen2 basics, (8) Power BI common gotchas and pitfalls (bidirectional filtering, AutoExist, blank-row), (9) workspace identity and OAuth2 / service-principal auth, (10) semantic model architecture review. Provides: star-schema templates, mode-selection matrix, incremental refresh recipe, gateway setup steps, and a common-gotchas reference.
Build messaging agents and apps with Spectrum — Photon's unified messaging SDK. Write your handler logic once and ship it across iMessage, WhatsApp Business, the terminal, or a custom platform. Spectrum is multi-platform by design and is becoming multi-language; the current SDK is `spectrum-ts` (TypeScript), with additional language SDKs planned. Use this skill for any Spectrum question — quickstart, multi-platform setup, receiving messages, content builders, spaces and users, reactions and replies, platform narrowing, the built-in providers (iMessage cloud/local/dedicated with message effects, Terminal TUI test harness, WhatsApp Business 1:1), custom event streams, graceful shutdown, building your own provider with `definePlatform`, and the production architecture patterns Photon uses internally to ship agents that live natively inside IM apps (five-stage inbound pipeline with debounce → batch flush → mark as read → generate → send, in-flight cancellation with abort signals, drain-in-handler, carry-forward, idempotent retries via stable client GUIDs and a startIndex resume cursor, per-resource memory scope `resourceId` vs `threadId`, durable job-failure audit log). This is the entry point for the skill; consult the topic files in this directory for full reference. Keywords: spectrum, spectrum-ts, photon, unified messaging, multi-platform, multi-language, im agent, messaging agent, imessage, whatsapp, whatsapp business, terminal, tuichat, definePlatform, custom platform, platform provider, platform narrowing, app.messages, Spectrum(), space, send, reply, react, tapback, typing indicator, responding, startTyping, stopTyping, content builder, text, attachment, voice, contact, richlink, poll, group, custom content, message effects, bubble effect, screen effect, line model, dedicated line, shared pool, custom events, app.stop, lifecycle, SIGINT, graceful shutdown, message queue, debounce, batch, in-flight, cancellation, abort controller, carry forward, idempotent retry, client guid, dedup, deduplication, startIndex, resume cursor, working memory, resourceId, threadId, per-resource memory, job failure, audit log, race condition, worker crash, retry, pg-boss, queue worker, conversational agent, chat agent, native messaging, agent architecture, production agent, spectrum patterns, best practices.
Autonomous iterative research loop. Takes a topic, runs web searches, fetches sources, synthesizes findings, and files everything into the wiki as structured pages. Based on Karpathy's autoresearch pattern: program.md configures objectives and constraints, the loop runs until depth is reached, output goes directly into the knowledge base. Triggers on: "/autoresearch", "autoresearch", "research [topic]", "deep dive into [topic]", "investigate [topic]", "find everything about [topic]", "research and file", "go research", "build a wiki on".
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.**
BEVFusion for multi-sensor 3D object detection. Fuses LiDAR point clouds and camera images in bird's-eye-view (BEV) space, used in autonomous driving for robust 3D perception. Use when training, evaluating, or running inference for a TAO BEVFusion model. Trigger phrases include "train BEVFusion", "LiDAR + camera fusion", "BEV 3D detection", "multi-sensor 3D perception".
Install cuOpt for Python, C, or server via pip, conda, or Docker; verify the install. For building cuOpt from source, see cuopt-developer.
High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.
Laboratory automation toolkit for controlling liquid handlers, plate readers, pumps, heater shakers, incubators, centrifuges, and analytical equipment. Use this skill when automating laboratory workflows, programming liquid handling robots (Hamilton STAR, Opentrons OT-2, Tecan EVO), integrating lab equipment, managing deck layouts and resources (plates, tips, containers), reading plates, or creating reproducible laboratory protocols. Applicable for both simulated protocols and physical hardware control.
Searching internet for technical documentation using llms.txt standard, GitHub repositories via Repomix, and parallel exploration. Use when user needs: (1) Latest documentation for libraries/frameworks, (2) Documentation in llms.txt format, (3) GitHub repository analysis, (4) Documentation without direct llms.txt support, (5) Multiple documentation sources in parallel