ScientiaCapital

Relationship intelligence graph for GTM work. Reads and writes Contact/Account/Deal/Outcome nodes in Neo4j Aura. Syncs contacts/accounts from HubSpot MCP and imports call outcomes from Nooks MCP automatically. Surfaces what messaging, sequences, and personas worked across verticals. Use when: relationship graph, gtm brain, log outcome, what worked with, account map, sync from hubspot, import nooks calls, contact history, sequence performance.

Live firm-wide sales pulse from the Epiphan CRM — revenue vs pace, pipeline by stage, won/lost, BDR activity, with coaching takeaways. Use when: business pulse, how are we doing, pipeline health, revenue pace, weekly numbers, standup brief, are we on track, sales snapshot.

Identify stalled and dying deals, run structured disqualification or recovery workflows, and clean pipeline of dead weight. Finds deals stuck 60+ days, with no champion, no activity, or overdue close dates and prescribes either a final re-engagement campaign or formal disqualification. Keeps pipeline honest and frees mental bandwidth. Use when: 'dead deals', 'stalled deals', 'clean pipeline', 'pipeline cleanup', 'disqualify deals', 'deal graveyard', 'zombie deals', 'which deals should I kill', 'pipeline hygiene', 'deal recovery', 'win back', 'lost deal review', 'close out stale deals'.

Chris Orlob's 5-step discovery framework applied to Epiphan — business problem → cause analysis → negative impact → future state → close. Per-vertical root causes, calibrated question bank, MEDDIC integration, and the 0-3 discovery scoring rubric used by sdr-call-coaching. Use when: discovery framework, orlob discovery, discovery questions, cause analysis, business problem discovery, how to run discovery.

Epiphan Video B2B sales - video capture/streaming lead qualification, pipeline metrics, MEDDIC discovery, and demo execution for Pearl devices, EC20 PTZ, and Epiphan Connect. Use for lead scoring, cold outreach to Higher Ed/Government/Corporate AV, and pipeline reviews.

5-touch email/SMS cadence for Epiphan BDR outreach. Full templates for Higher Ed, Courts, Gov, Healthcare by touch. Companion to epiphan-call-playbook (phone). Writing rules, subject line patterns, and SMS guidelines. Use when: email playbook, email template, outreach email, sdr email, cold email template, sms outreach, follow up email, 5 touch cadence.

Review and batch-send staged Gmail BDR outreach drafts. Lists all queued drafts, shows a prioritized review table (recipient / subject / vertical), and generates one-click Gmail send links so Tim can flush the outreach queue in under 2 minutes. Trigger: send staged drafts, flush draft queue, batch send drafts, review my drafts, what drafts do I have.

Friday 5:30 PM CST weekly KPI report for Tim's SDR pod (Tim Kipper, Edgar Marroquin, Vasil Ivanov). Pulls each rep's real Nooks dials/connects + HubSpot meetings-booked and pipeline-created for Mon 00:00 → Fri 5:30 PM CST, calculates connect rate + attainment vs weekly targets, builds color-coded per-rep scorecards (GREEN/YELLOW/RED) with week-over-week trend arrows, rolls up a team totals row, and delivers an HTML report + a compact Slack DM to Tim. Read-only everywhere except the Slack DM and the local HTML/prev-week JSON. This task MEASURES the reps; sdr-call-coaching (Fridays 11 AM) COACHES them — do not duplicate the coaching rubric here.

Self-learning workflow system that tracks what works best for your use cases. Records experiment results, suggests optimizations, creates custom templates, and builds a personal knowledge base. Use to learn from experience and optimize your LLM workflows over time.

Create, clean, and optimize datasets for LLM fine-tuning. Covers formats (Alpaca, ShareGPT, ChatML), synthetic data generation, quality assessment, and augmentation. Use when preparing data for training.

Export and deploy fine-tuned models to production. Covers GGUF/Ollama, vLLM, HuggingFace Hub, Docker, quantization, and platform selection. Use after fine-tuning when you need to deploy models efficiently.

Advanced techniques for optimizing LLM fine-tuning. Covers learning rates, LoRA configuration, batch sizes, gradient strategies, hyperparameter tuning, and monitoring. Use when fine-tuning models for best performance.

Train and use SuperBPE tokenizers for 20-33% token reduction across any project. Covers training, optimization, validation, and integration with any LLM framework. Use when you need efficient tokenization, want to reduce API costs, or maximize context windows.

Analyze, compare, and work with tokenizers using Unsloth tools. Compare different tokenizers, analyze token efficiency, and integrate with Unsloth models. For SuperBPE training, see the 'superbpe' skill.

Fine-tune LLMs 2x faster with 80% less memory using Unsloth. Use when the user wants to fine-tune models like Llama, Mistral, Phi, or Gemma. Handles model loading, LoRA configuration, training, and model export.

Work with the Unsloth MCP Server codebase. Use when maintaining, extending, or debugging this specific MCP server implementation. Provides architecture knowledge, code patterns, and development workflows.

Implement cost-optimized LLM routing with NO OpenAI. Use tiered model selection (DeepSeek, Haiku, Sonnet) to achieve 70-90% cost savings. Triggers on "LLM costs", "model selection", "cost optimization", "which model", "DeepSeek", "Claude pricing", "reduce AI costs".

Deploy projects to Vercel, Railway, or Docker with platform-specific best practices. Use when deploying applications, configuring deployment settings, debugging deployment failures, or setting up CI/CD pipelines. Triggers on "deploy to vercel", "railway deployment", "docker build", "deployment failed", "configure vercel.json".

GTM workflows for revenue acceleration across Scientia projects. Use for demo preparation, sales outreach, battle cards, pricing strategy, and revenue tracking. Triggers on "revenue focus", "prepare demo", "sales outreach", "battle card", "GTM strategy", "pricing", "tier-1 projects".

Master Supabase patterns for migrations, RLS policies, pgvector, and authentication. Use when creating database schemas, writing migrations, implementing row-level security, setting up auth, or debugging Supabase issues. Triggers on "supabase migration", "RLS policy", "row level security", "pgvector", "supabase auth", "magic link".

Generate new Claude Skills for the Dual-Domain LLM Platform using natural language. Use when the user wants to create a new skill, automate a workflow, or build team capabilities. Activates when user mentions "create a skill", "new skill", "skill for", or "automate [task]".

Save 40-70% on AI costs with intelligent multi-LLM routing. Automatically selects the optimal model based on task complexity across 40+ models from 8 providers.