Search: "does"

Performs RNA velocity analysis with scVelo from spliced and unspliced single-cell RNA counts. Fits deterministic or dynamical models, examines gene phase portraits, builds velocity graphs, estimates relative latent time, and ranks velocity-associated genes. Use for directional trajectory hypotheses and kinetic-model diagnostics alongside Scanpy; velocity alone does not establish cell fate or causal drivers.

Check code against the documentation that specifies it - which requirements hold, which the code contradicts, which are absent, and what the code does that no document mentions. Use when comparing an implementation against a whitepaper, protocol spec, or design document.

Conducts systematic, scoping, and narrative literature reviews using PubMed, arXiv, bioRxiv, Semantic Scholar, and other appropriate sources. Use for research synthesis, reproducible literature searches, screening, citation checking, or preparing Markdown and PDF reviews. Tracks search coverage, records versus studies, and evidence limitations; supports meta-analysis planning but does not supply a meta-analysis engine.

GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster. Use for CUDA/GPU optimization; CPU-bound NumPy, SciPy, pandas, scikit-learn, NetworkX, scikit-image, vector-search, image-processing, graph, simulation, or file-I/O workloads; CuPy, cuDF, cuML, cuGraph, cuVS, cuCIM, KvikIO, Warp, Newton, Numba-CUDA, or RAFT questions; and profiling, memory-transfer, kernel, or multi-GPU bottlenecks. Also use when large data-parallel Python code is slow and GPU acceleration is a plausible option, even if the user does not name CUDA.

Converts neuroscience acquisition data to Neurodata Without Borders files with NeuroConv and PyNWB, preserves metadata and timebases, checks evidence-based clock alignment, and produces schema validation, NWB Inspector findings and round-trip checks. Use for NWB conversion and synchronization of planar single-channel two-photon TIFF imaging plus timestamped behavioral position CSV; this skill does not perform spike sorting or claim tested support for arbitrary acquisition formats.

The AI that really does things. Any OS. Any Platform. The lobster way. 🦞

Use this skill when a publisher wants a multi-period earnings trend across their configured networks, how concentrated their earnings are in one or two programmes, and how long money typically takes to move from converted to approved to paid so they can forecast the next payout. This is the premium extension of the free `affiliate-earnings-report` skill, not a replacement for it. Trigger on: "Show my earnings trend over the last six months", "How concentrated is my income in one programme?", "How long does [network] usually take to pay me?", "Forecast when my pending commissions will land".

Use this skill when an agency operator wants the list of partner applications waiting on a decision for a brand: who has applied to the programme and is sitting in the pending queue. The output is a read-only queue; the approve or decline happens in the network dashboard. How the queue is read depends on the network. Some advertiser networks expose application status through the API. Awin does not: its advertiser API returns only joined publishers with no application status, so Awin's pending queue is read from the operator's own authenticated Awin session (new UI, app.awin.com) through Claude-in-Chrome, read-only. Awin Classic (ui.awin.com) accounts have no equivalent and are detected and stopped. Trigger on: "Which partner applications are pending on Acme?", "Acme's application queue", "Who's waiting for approval on [brand]?", "Show me pending partners for Acme".

Use this skill when a publisher or agency wants to find brands on Awin they have not yet joined and decide which ones to apply to next, ranked by fit. Discovery only: it reads the joinable programmes from the Awin API and produces a prioritised shortlist with reasons. It does not apply to anything. Trigger on: "Which brands should I apply to on Awin?", "Find new programmes to join", "Build my Awin application shortlist", "What programmes can I still join?", "Prioritise brands to apply for".

Premium pack. Benchmark one brand's affiliate programme against the other brands you manage in the same vertical: commission competitiveness, partner mix, reversal rate, and AOV, with a gap list the account manager can act on. Trigger on: "benchmark Acme against its vertical", "how does this brand compare to peers we manage", "is our commission competitive for this category".

A Chrome extension that reads the LinkedIn job posting you're viewing and gives you one-click searches for the people behind it: recruiters, the hiring team, past interns, and people posting about the role. It does not scrape people results or automate LinkedIn. It only reads the posting on scree...

Use this skill when an agency operator wants the partner roster for a brand split by relationship status, plus the active partners who have gone quiet so they can be chased or reactivated. The output is a read-only worklist; the outreach itself happens outside this skill. How a partner's relationship status is read depends on the network. Some advertiser networks report it through the API. Awin does not: its advertiser API returns only already-joined publishers with no relationship-status field, so every Awin row maps to unknown and neither the roster split nor the dormant worklist can be derived from the API. Awin's roster and per-partner relationship status are read from the operator's own authenticated Awin session (new UI, app.awin.com) through Claude-in-Chrome, read-only. Awin Classic (ui.awin.com) accounts have no equivalent and are detected and stopped. Trigger on: "Who's gone quiet on Acme?", "Acme's partner roster health", "Which partners have gone dormant on [brand]?", "Build a reactivation list for Acme", "Show me Acme's roster by status".