Menu
Skills in this repository
SkillsMP has collected 924 skills from mcorbett51090/RavenClaude. Open a skill to review its source and details.
mcorbett51090/RavenClaudeShowing 40 of 924 collected skills.
Run a repeatable brand-standard audit across franchise units and turn it into a coaching loop: a weighted checklist tied to the franchise agreement's operating standards, a scored audit per unit, the specific gaps framed as contract risk, and a…
Design, calibrate, and bias-audit an LLM-as-judge so its scores track human judgment: named criteria with anchored levels (not 'rate 1-10'), a structured verdict + reason, a pinned judge model/version, a human-agreement calibration on a sample, and checks for…
Turn a fuzzy 'is the AI good enough?' into a concrete measurement plan: task spec, the eval method from the decision tree (exact / rule-based / model-graded / human), the metric + what it misses, the offline-vs-online split, a power-adequate sample size, and…
Wire an offline eval suite into a CI ship-gate so a prompt/model change is scored against a frozen baseline and blocked on a regression past a preset threshold, with guardrail/red-team checks as a zero-tolerance blocker and cost/latency tracked alongside…
Engineer privacy: build executable data-subject-rights pipelines (access/erasure/portability) that locate data via the catalog, track lawful basis + granular revocable consent, minimize collection, automate retention/deletion, and distinguish pseudonymization…
Reversibly pseudonymize names/entities in text BEFORE sending it to a model, and restore them in the reply. Deterministic denylist (reliable) + structured PII + optional NER; local vault; residual scan. Pseudonymizes, does NOT anonymize.
Wire a live GL/accounting-system OAuth2 extractor (QuickBooks Online / NetSuite / Sage Intacct / Xero) that feeds the canonical trial-balance staging seam, using the reference-implementation token client in scripts/connectors/ — atomic persist-then-use…
Build a non-React static site: semantic HTML, token-driven CSS (custom properties, grid/flex, container queries, logical properties), and an SSG-first Astro/11ty/Hugo or plain HTML+CSS build with islands only where earned. Covers reflow to 320px, in-markup…
The governed review→approve→lock state machine with enforced segregation of duties and an append-only hash-chained audit log. Refuses same-actor approval above threshold and illegal transitions. Runs scripts/close_state.py. Honest tier caveats. Used by…
Produce the recurring close supporting schedules — fixed-asset depreciation rollforward, prepaid amortization, and deferred-revenue waterfall — from summarized inputs, each self-checking so beginning + movements == ending (and NBV = cost − accumulated…
Normalize a raw accounting-system trial-balance export (QuickBooks Online / NetSuite / Xero / Sage Intacct) into the ONE canonical staging schema — account,description,debit,credit,entity,period,currency — that the close autopilot consumes, via a data-driven…
Render a close-package JSON (controller_cycle.py --out-json) into a self-contained, CSP-safe, theme-aware per-entity dashboard — headline KPIs (revenue, net income, gross margin, current ratio, DSO), IS/BS summaries, reconciliation exceptions, top flux, and…
Auto-match GL journal lines to sub-ledger / bank lines (exact, tolerance, and many-to-one grouped) and AUTO-CERTIFY an account only when its unexplained residual is within materiality — else FLAG for a human. The auto-match engine reconcile_summary.py's…
Balance-sheet tie-out (book vs sub-ledger, review-by-exception at materiality) plus a materiality-suppressed period-over-period flux table. Runs scripts/reconcile_summary.py; narrative 'why' reuses the variance-commentary skill. Used by `controller`.
Decide which revenue lines a creator should run and in what order — grounded in audience size × engagement × buying-intent, weighted toward recurring revenue and away from single-platform/single-sponsor concentration. Reach for this at the START of monetizing…
Plan a creator's content strategy, sustainable cadence, platform growth, and owned-audience funnel: content pillars, a capacity-matched schedule, repurposing one core piece into platform-native cuts, and converting rented reach into an email list/community —…
Value a sponsorship and build a creator rate card on VALUE, not a flat CPM: audience value + deliverable + usage/exclusivity rights as separate line items + a walk-away number, with rate benchmarks dated + verify-at-use and clear disclosure. Reach for this…
Review a schematic and PCB layout before fab: correctness (ERC/DRC, footprints/pinouts/polarity, power), integrity (decoupling at the pin, grounding/return paths, stack-up, controlled impedance), and DFM/DFA/DFT against the target fab's rules — returning a…
Turn a product idea into a hardware architecture: decide module/dev-board vs custom PCB (via the decision tree), name requirements before parts, sketch the power tree, the interface plan, and an EMC/pre-compliance posture — against a…
Select the MCU/SoC and key components and build a supply-aware BOM: match parts to requirements, read datasheet parameters at the operating point (worst-case + margin), and weigh availability / second-source / lifecycle / cost at the target volume. Reach for…
Decide whether to federate, keep a monolithic graph, or stitch — and if federating, design clean subgraph ownership boundaries with entity @key, reference resolvers, and @external/@requires/@provides, then weigh the real operational cost of running a…
Design a GraphQL schema for the clients that consume it — not the database underneath: nullability discipline, Relay cursor-connection pagination, input/payload mutation shape, errors-as-data vs top-level errors — and evolve it without breaking clients via…
Defend a GraphQL endpoint whose single URL accepts arbitrary nested queries: depth limiting, cost/complexity budgets, field-level authorization (authorize at the field, not the endpoint), introspection hardening in prod, persisted/allow-listed operations,…
Kill the N+1 problem GraphQL's per-field resolution invites: batch and per-request-cache with DataLoader, be selection-set aware so you fetch only requested fields, bound pagination cost, avoid over-fetching downstream, and layer response caching / APQ.…
Build the recurring-revenue engine: design club/membership tiers on member lifetime value (shipment value/frequency, benefits), read and reduce churn by cohort, and manage DTC e-commerce. A club member is worth more than a case sold — but churn quietly eats…
Cost a craft-beverage unit and plan production against real capacity: decompose COGS per unit (raw material, yield loss, packaging, overhead absorption), read tank/barrel/time capacity and turns, and plan batches against the demand-by-channel plan. You can't…
Read and lift the tasting-room funnel: visits -> tasting -> purchase -> club sign-up. Find where conversion leaks, tie it to the experience and the club offer rather than raw foot traffic, and weigh tasting room vs e-commerce vs events on contribution. DTC…
Model the go-to-market structure: self-distribution (margin kept, sales/logistics cost, eligibility limits) vs a distributor (reach and depletion, margin given away, franchise-law lock-in), plus the channel margin math (DTC net vs wholesale net after…
Turn aesthetic consults into booked treatments: read where conversion leaks (offer clarity, provider-set treatment plan, transparent pricing, financing, same-day booking, follow-up cadence) and fix the highest-leak step first — without making a clinical or…
Map the med-spa compliance structure — scope of practice, good-faith exam / medical supervision, consent and adverse-event protocols, product handling — as operational structure, flagging every state-specific rule [verify-at-use] and routing the actual…
Tilt the med-spa service mix toward contribution: read revenue and margin per provider-hour and per room-hour across injectables, energy devices, skincare retail, and memberships; model capital-device payback on realistic volume; size memberships on…
Read a med spa on its two scarce inventories — injector-hours and treatment-room-hours. Utilization = productive hours booked / available, read by provider and by room and by daypart; confirm the current capacity is full before adding an injector, a room, or…
Curate and augment the dataset, choose and fine-tune the model (YOLO / DETR / SAM / CLIP / EfficientNet / ViT) on the task and target budget, design the loss and metric to match the cost, spend the annotation budget with active learning, handle class…
Frame the computer-vision task (classification / detection / segmentation / OCR / pose / tracking / VLM) on the decision the system must make, choose the metric that mirrors the business cost, decide build-vs-fine-tune-vs-API jointly with the deployment…
Design streaming-video pipelines that hold the frame budget — frame sampling / keyframe strategy, ROI cropping, tracking-by-detection so the detector doesn't run every frame, and batching where latency allows — and deploy to edge/embedded targets (Jetson,…
Make a trained vision model fit and run on its target: budget latency on the real device, optimize in order of leverage (quantization INT8/FP16 with calibration, then pruning, then distillation), export to the runtime the target uses (ONNX / TensorRT / CoreML…
Design the dialog layer of a voice agent: explicit dialog-state modeling, LLM orchestration, context/window management over a long call, and mid-call function/tool calling that hides latency with natural fillers and treats tool failure as a first-class…
Engineer the ASR and TTS layer of a voice agent: choose STT/TTS providers on the channel (narrowband telephony vs wideband web), languages, streaming support, WER, and cost; stream transcription and synthesis to cut latency; tune VAD/endpointing; handle…
Integrate a voice agent with real telephony and validate it: SIP/PSTN trunking and media handling, WebRTC for web/app, DTMF capture (in-band vs RFC 2833/SIP INFO), IVR-vs-conversational per step, call routing and warm/cold human transfer with context, and a…
Choose the voice-agent pipeline shape (cascade STT->LLM->TTS vs speech-to-speech), the channel (telephony / web / SDK), and the build-vs-platform bet (Twilio / Vapi / Retell / LiveKit / Pipecat), then allocate the end-to-end latency budget per hop and design…