JeromyJSmith

Use when user asks to 'lint agent configs', 'validate skills', 'check CLAUDE.md', 'validate hooks', 'lint MCP'. Validates agent configuration files against 385 rules.

Interprets Culture Index (CI) surveys, behavioral profiles, and personality assessment data. Supports individual profile interpretation, team composition analysis (gas/brake/glue), burnout detection, profile comparison, hiring profiles, manager coaching, interview transcript analysis for trait prediction, candidate debrief, onboarding planning, and conflict mediation. Accepts extracted JSON or PDF input via OpenCV extraction script.

Creates devcontainers with Claude Code, language-specific tooling (Python/Node/Rust/Go), and persistent volumes. Use when adding devcontainer support to a project, setting up isolated development environments, or configuring sandboxed Claude Code workspaces.

Analyzes smart contract codebases to identify state-changing entry points for security auditing. Detects externally callable functions that modify state, categorizes them by access level (public, admin, role-restricted, contract-only), and generates structured audit reports. Excludes view/pure/read-only functions. Use when auditing smart contracts (Solidity, Vyper, Solana/Rust, Move, TON, CosmWasm) or when asked to find entry points, audit flows, external functions, access control patterns, or privileged operations.

Draws 4 Tarot cards to inject entropy into planning when prompts are vague, ambiguous, or casually delegated. Interprets the spread to guide next steps. Use when the user says 'let fate decide', 'YOLO', 'whatever', 'idk', or other nonchalant phrases, makes Yu-Gi-Oh references, or when you are about to arbitrarily pick between multiple reasonable approaches. Prefer over ask-questions-if-underspecified when the user's tone is casual or playful rather than precision-seeking.

Configures mewt or muton mutation testing campaigns — scopes targets, tunes timeouts, and optimizes long-running runs. Use when the user mentions mewt, muton, mutation testing, or wants to configure or optimize a mutation testing campaign.

Runs external LLM code reviews (OpenAI Codex or Google Gemini CLI) on uncommitted changes, branch diffs, or specific commits. Use when the user asks for a second opinion, external review, codex review, gemini review, or mentions /second-opinion.

Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language.

Reproductive function. Copies the root .Codex/skills/ genome into a target section's .Codex/skills/ location (or deeper), appends the parent commit SHA to each child SKILL.md lineage, and registers the new cell in lattice/knowledge/skills_registry. INERT in Phase 1 — do not execute recursive division yet.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

LATTICE codebase graph introduction — communities, hotspots, dependency paths for VW_iTwin_Bridge

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

Compact the current conversation into a handoff document for another agent to pick up.

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

Work inside a running marimo notebook's kernel — execute code, create cells, and build a notebook as an artifact. Use when the user wants to start a marimo notebook or work in an active marimo session.

Introspective function. The organism inspects its own codebase via search_tutorials, search_research, search_docs, search_api_reference, get_coverage_gaps, Graphify dependency graphs, and InfraNodus topic analysis, then proposes a polymorphic decomposition plan. NEVER executes cell-divide. Human gate is mandatory. Depth-guarded: max 3 new levels of nesting per cycle. INERT until Phase B.

Generates personalized outreach scripts for motivated sellers across any asset class. Profiles seller type — retiree, estate/probate, tired landlord, distressed, institutional — and crafts phone scripts, email drafts, and follow-up sequences. Scripts are built around seller pain points, not buyer wants. Use when user says "write outreach", "generate scripts", "/outreach", "how do I contact this seller", "write me a pitch", "create seller scripts", "what do I say to this broker", or "write a cold call script".

Full 6-stage deal intelligence pipeline for finding motivated-seller opportunities across any asset class. Scans CREXI, auctions, BizBuySell, and Zillow, applies regional market intelligence and prediction market signals, detects hidden gems, and produces ranked HOT/WARM/WATCH deals with outreach scripts. Use when user says "run hunt", "find deals", "scan for opportunities", "/hunt", "run the pipeline", "find me deals in Nevada County", "what deals are available today", "run all zones", or "start the deal pipeline".

Generates formatted deal intelligence reports from pipeline output. Produces executive summary, zone-by-zone breakdown for NCA (Grass Valley/Nevada City CA), Colorado, and nationwide top 3, hidden gems section, prediction market context, and ready-to-send deal packages. Use when user says "generate report", "/report", "summarize deals", "create deal summary", "what did the pipeline find", "show me the results", or "compile the findings".

Multi-source deal scanner. Scrapes CREXI, Auction.com, Hubzu, BizBuySell, and Zillow for motivated-seller opportunities across any asset class — real estate, businesses, land, notes. Respects rate limits. Saves structured JSON to data/raw/. Use when user says "scrape listings", "scan for deals", "/scrape", "search [location] for deals", "check auction.com", "find owner-financed listings", "search BizBuySell", or "get raw listings from [source]".

Scores and ranks deal opportunities using a 6-category rubric covering seller motivation, deal economics, financing terms, regional intelligence, prediction market signals, and asset quality. Detects hidden gems — deals that look bad on the surface but are exceptional when regional and macro context is applied. Classifies deals as HOT, WARM, WATCH, or PASS. Use when user says "score deals", "rank opportunities", "/score", "which deals are hot", "find the best deals", "rank what we found", or "score this listing".

Sets up and manages daily automated deal monitoring. Configures cron schedule to run the full pipeline daily, tracks new deals since last run, surfaces only net-new opportunities, and maintains a deal changelog. Use when user says "set up daily monitoring", "/watch", "run this every day", "automate the hunt", "monitor for new deals", "schedule daily scans", "set up alerts", or "run pipeline on a schedule".

Fetches real-time probability data from Kalshi and Polymarket to generate deal timing and acquisition strategy signals. Translates Fed rate decisions, recession odds, and regional housing index probabilities into actionable recommendations. Use when user says "check prediction markets", "what do markets say", "get macro signals", "should I act now or wait", "what are the odds", "check Kalshi", "check Polymarket", "what is the market timing", or "give me the economic outlook".

Profiles motivated sellers to determine pain points, urgency level, and optimal negotiation approach. Identifies seller type from listing language — retiree, estate/probate, tired landlord, distressed, institutional — across any asset class. Use when user says "profile this seller", "analyze seller motivation", "who is the seller", "how motivated are they", "what is the seller situation", "how should I approach this deal", "what is driving this listing", or "rate seller motivation".

Claude Code hooks patterns. Use when writing hook scripts (PreToolUse, PostToolUse, Stop, SessionStart, etc.) or configuring .claude/settings.json to wire hooks. Reference lib/claude-code-hooks-mastery/ for working examples.

Interpretable Context Methodology — folder structure as agent architecture. Use when creating new stages, reading stage CONTEXT.md files, or understanding how this workspace routes work across numbered stage folders. Reference lib/ICM/ and lib/ICM-animation-example/ for working implementations.

just-prompt MCP server for model-agnostic prompt routing. Use when you need to send the same prompt to multiple LLM providers, compare outputs, or route prompts based on provider/model selection. Installed via uv sync.

Multi-agent observability patterns. Use when adding logging, tracing, or monitoring across agent interactions. Reference lib/claude-code-hooks-multi-agent-observability/ for working examples including agent session tracking, damage control, and multi-agent event streams.

Pi Coding Agent extension playground. Use when writing, running, or debugging Pi extensions for peer-to-peer communication (coms-net, coms, pi-pi) or multi-agent workflows (agent-team, agent-chain). Bun only — never pnpm/npm.

Development skill for the Global Rome memorial app — a collective resonance web application for Rome Viharo (1967-2025). Use when building, modifying, or debugging the Global Rome app (Next.js 16 + Tone.js + Three.js + Supabase + Neo4j + Polygon). Covers frequency-to-color synesthesia mapping, triangle spin mechanics, resonance detection, Rome Score calculation, audio synthesis (formant/NOW sweep), ceremony mode projection, Neo4j knowledge graph operations, and Supabase realtime broadcasting. Reference this skill for any work on the memorial app codebase in global-ohm/ directory.

Reference codebase for Global Ohm. Use this skill when you need to understand the structure, implementation patterns, or code details of the Global Ohm project.

Direct browser control via CDP. Use when the user wants to automate, scrape, test, or interact with web pages. Connects to the user's already-running Chrome.

Turborepo monorepo build system guidance. Triggers on: turbo.json, task pipelines, dependsOn, caching, remote cache, the "turbo" CLI, --filter, --affected, CI optimization, environment variables, internal packages, monorepo structure/best practices, and boundaries. Use when user: configures tasks/workflows/pipelines, creates packages, sets up monorepo, shares code between apps, runs changed/affected packages, debugs cache, or has apps/packages directories.