joemccann

@joemccann source repo

36 skills
0 followers
3 hours ago last updated

▌
Frontend Design · joemccann bundle
Design thinking and decision-making for web UI. Use when designing components, layouts, color schemes, typography, or creating aesthetic interfaces. Teaches principles, not fixed values.

0
installs
▌
CI Performance 2 · joemccann
Nightly CI and deploy optimizer - measure the real push-to-green-production critical path from GitHub Actions and production deploy timestamps, then land every ranked, evidence-backed optimization that passes the rails on the dated PR branch (one commit per experiment so each stays attributable) without weakening any test, gate, provenance, health, recovery or rollback guarantee, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/ci_performance_nightly.sh, one daily cycle at 00:20 local that runs audit, remediate, then deliver; invoke as /ci-performance audit, /ci-performance remediate or /ci-performance deliver.

0
installs
▌
Testing Weekend 2 · joemccann
Weekend testing loop - daily delta-audit of test-suite health for everything merged since the last audited SHA (new findings appended to TEST_AUDIT.md), then red/green remediation of EVERY verified finding on the dated PR branch, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/testing_weekend.sh, one daily cycle at 00:10 local that runs audit, remediate, then deliver; invoke as /testing-weekend audit, /testing-weekend remediate or /testing-weekend deliver.

0
installs
▌
Reliability Weekend 2 · joemccann
Weekend reliability loop - daily delta-audit of everything merged since the last audited SHA (new findings appended to RELIABILITY_AUDIT.md), then red/green remediation of EVERY verified finding on the dated PR branch, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/reliability_weekend.sh, one daily cycle at 00:00 local that runs audit, remediate, then deliver; invoke as /reliability-weekend audit, /reliability-weekend remediate or /reliability-weekend deliver.

0
installs
▌
Documentation Nightly 2 · joemccann
Nightly documentation maintainer - daily audit that classifies the documentation impact of everything merged since the last audited SHA (rolling issue labeled documentation-nightly), then smallest source-backed remediation of EVERY verified P0/P1/P2 finding on the dated PR branch documentation/<date> without inventing prose, duplicating machine truth, or touching live systems, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/documentation_nightly.sh, one daily cycle at 00:30 local that runs audit, remediate, then deliver; invoke as /documentation-nightly audit, /documentation-nightly remediate or /documentation-nightly deliver.

0
installs
▌
Quant Backtest · joemccann bundle
Institutional-grade Python backtesting framework builder for Codex. Use this skill whenever the user mentions backtesting, quant strategy, alpha model, trading system, signal engine, portfolio backtest, walk-forward optimization, strategy performance, Sharpe ratio calculation, look-ahead bias, transaction cost modeling, or building any systematic trading infrastructure. Also trigger on mentions of vectorized signals, position sizing, mark-to-market, risk metrics (VaR, Sortino, Calmar), regime filters, or factor attribution. If the user says "backtest this idea" or "test my trading strategy", use this skill.

0
installs
▌
Canvas Design · joemccann bundle
Create beautiful visual art in .png and .pdf documents using design philosophy. Use this skill when the user asks to create a poster, piece of art, design, or other static visual piece. Creates original visual designs with expert-level craftsmanship, never copying existing artists' work to avoid copyright violations.

0
installs
▌
Exa Search · joemccann
Semantic web search, similar page discovery, and content extraction via Exa AI API. Use when the user needs meaning-based search (not keyword matching), wants to find pages similar to a URL, or needs clean content extraction from web pages. Requires EXA_API_KEY environment variable.

0
installs
▌
PDF · joemccann bundle
Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.

0
installs
▌
Xai Grok · joemccann
xAI Grok APIs for image generation, chat, vision, live search, and video. Use when the user asks to use Grok, xAI, or Grok Imagine for generating images, chatting with Grok, analyzing images with Grok Vision, searching the web/news/X with Grok, or generating videos. Triggers on "grok", "xai", "grok imagine", "ask grok", "grok vision", "grok search".

0
installs
▌
Stream Browser · joemccann
Start a streaming browser session with real-time visual feedback

0
installs
▌
Autonomous Browse · joemccann
Run an autonomous browser agent to complete a task on a website

0
installs
▌
Browser Screenshot · joemccann
Take a screenshot of a webpage using the streaming browser

0
installs
▌
Glimpse · joemccann bundle
Show native macOS UI from scripts and agents — dialogs, forms, visualizations, floating widgets, cursor companions. Use when you need to display HTML to the user, collect input, show a chart, render markdown, or create any visual interaction without a browser.

0
installs
▌
Swiftui UI Patterns · joemccann bundle
Best practices and example-driven guidance for building SwiftUI views and components. Use when creating or refactoring SwiftUI UI, designing tab architecture with TabView, composing screens, or needing component-specific patterns and examples.

0
installs
▌
Metal Macos Replatform · joemccann bundle
Use this skill when planning, auditing, or implementing Metal-backed macOS desktop surfaces, especially hybrid SwiftUI/AppKit plus MetalKit apps, shader packaging, MTKView integration, and Metal toolchain setup.

0
installs
▌
AI Chat UX Best Practices · joemccann bundle
Use this skill when planning, auditing, or refining the assistant UX in the native macOS client, especially setup, settings, chat trust surfaces, prompt wrappers, and command transparency.

0
installs
▌
Macos Branded Notifications · joemccann bundle
Send branded macOS notifications with custom app icon using terminal-notifier. Use when implementing native Notification Center integration with custom branding.

0
installs
▌
Menthorq · joemccann
MenthorQ data extraction — CTA positioning, dashboards, screeners, forex, summaries, and QUIN AI screener prompts. Triggers on "menthorq", "cta positioning", "menthorq dashboard", "menthorq screener", "quin screener", "menthorq forex", or any MenthorQ-related query.

0
installs
▌
Tweet It · joemccann bundle
Generate tweet copy and a matching infographic image for sharing trades on X (Twitter). Use when the user says "tweet-it", "tweet this trade", "create a tweet", "X post", or wants to share a trade on social media. Produces tweet text (ready to copy), an infographic card image, and a self-contained preview page with copy buttons for both text and image.

0
installs
▌
Web Fetch · joemccann
Web searching and content fetching. Use Exa MCP tools (web_search_exa, company_research_exa, get_code_context_exa) as the PRIMARY method for web search, company research, and content retrieval. Fall back to agent-browser only when Exa cannot handle the task (e.g., interactive forms, authentication, screenshots, JavaScript-rendered content). Triggers include "search the web", "find information about", "look up", "research this company", "fetch this page", "scrape data from a page", "open a website", or any task requiring web content.

0
installs
▌
HTML Report · joemccann bundle
Generate styled HTML reports using the Terminal Dashboard theme. Use when the user needs to create an HTML report, dashboard, data visualization, or styled output document. Triggers include requests to "generate a report", "create an HTML file", "make a dashboard", "visualize data", "export to HTML", or any task requiring formatted HTML output with the project's visual identity.

0
installs
▌
Indicator · joemccann
Parallel TDD swarm that ships a new market indicator end to end. Usage - /indicator <data-source> <indicator-name>. Researches the source, writes a spec plus failing tests, fans out three worktree implementer subagents (ingestion, API, chart tab), merges, runs the full suite, screenshots the live tab with Playwright, then commits, pushes, waits for CI green, and verifies production.

0
installs
▌
Options Analysis · joemccann
On-demand options pricing and structure analysis for the Radon trading agent

0
installs
▌
New Indicator · joemccann
The canonical Radon pattern for shipping a new market indicator end to end (ingestion job, Turso storage, API route, regime chart tab, tests, systemd timer, CI, production verify). Derived from the BREADTH (21c13715, f067491e, 770d4674) and FINRA margin-debt (c5f4b877, b2f452fb, 189fcad8) builds. Read before adding any indicator; /indicator orchestrates a parallel swarm on top of this pattern.

0
installs
▌
CI Performance · joemccann bundle
Nightly CI and deploy optimizer - measure the real push-to-green-production critical path from GitHub Actions and production deploy timestamps, then land every ranked, evidence-backed optimization that passes the rails on the dated PR branch (one commit per experiment so each stays attributable) without weakening any test, gate, provenance, health, recovery or rollback guarantee, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/ci_performance_nightly.sh, one daily cycle at 00:20 local that runs audit, remediate, then deliver; invoke as /ci-performance audit, /ci-performance remediate or /ci-performance deliver.

0
installs
▌
Browser Use Cloud · joemccann
Cloud-based AI browser agent for autonomous web tasks. Use when you need to perform complex browser automation tasks autonomously, scrape websites that require AI reasoning, fill forms intelligently, extract structured data from web pages, or perform multi-step web workflows. Triggers include "autonomous browsing", "AI browser", "browser-use", "scrape with AI", "extract structured data from website", "fill out form intelligently", "navigate website and find", or when the task requires reasoning about page content rather than just clicking elements. Prefer this over web-fetch when the task involves multiple steps, requires understanding page semantics, or when you need structured output.

0
installs
▌
Fast Regex Search · joemccann
Trigram-indexed regex search for large codebases. Loads automatically on startup to accelerate all grep/search operations. Implements sparse n-gram indexing with frequency-weighted trigram decomposition, bloom-filter adjacency masks, and mmap'd posting lists for sub-50ms regex matching across 500k+ files.

0
installs
▌
Testing Weekend · joemccann bundle
Weekend testing loop - daily delta-audit of test-suite health for everything merged since the last audited SHA (new findings appended to TEST_AUDIT.md), then red/green remediation of EVERY verified finding on the dated PR branch, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/testing_weekend.sh, one daily cycle at 00:10 local that runs audit, remediate, then deliver; invoke as /testing-weekend audit, /testing-weekend remediate or /testing-weekend deliver.

0
installs
▌
Ib Order Execution · joemccann
Execute and monitor options orders via Interactive Brokers. Use when placing trades, monitoring fills, or managing orders. Triggers on "execute order", "place trade", "buy calls", "sell puts", "bull call spread", "bear put spread", "monitor fills", "check order status", or any IB order-related task.

0
installs
▌
Context Engineering · joemccann bundle
Implement file-system-based context engineering for agentic AI projects. Use when building persistent memory systems, context pipelines, token-budget management, context repositories, or agentic file system abstractions. Triggers on "context engineering", "context pipeline", "context management", "persistent memory", "token budget", "context constructor", "context evaluator", "agentic file system", "context rot", "memory lifecycle", or any task involving structured management of LLM context windows, history/memory/scratchpad tiers, or governance of context artifacts across agent sessions. Also use when asked to add memory, traceability, or context governance to an existing agentic project.

0
installs
▌
Security Nightly · joemccann
Nightly security auditor and authorized local penetration tester - daily audit that scans the source delta since the last audited SHA with pinned deterministic tools plus harvest of whatever Vercel DeepSec sibling export is already ready and the official Claude Security plugin, independently verifies every candidate against current code, then remediates every independently verified source-actionable finding with a durable regression, then a deliver phase that pushes P2/P3 (and operator-released P0/P1) fixes as one sanitized PR, gets CI green and tells the operator what to merge. DeepSec is a sibling worker (scripts/security_deepsec_worker.sh, own launchd/cap/dead-man), not a second remediate/deliver loop. Runs unattended and CREDENTIAL-FREE in ~/radon-weekend/radon-security via scripts/security_nightly.sh, one daily cycle at 00:40 local (audit, remediate, then deliver); invoke as /security-nightly audit, /security-nightly remediate or /security-nightly deliver. Fails closed and never touches production, live

0
installs
▌
Incident Response · joemccann
The exact Radon triage playbook for a production incident - reproduce locally, write a FAILING regression test first, fix, run the full test gate, commit a focused single-purpose commit, push a fix/** branch, ensure an open PR (never merge), wait for CI green, verify live in the browser. Use whenever an incident JSON from data/incidents/ or a production bug report is being worked; /incident orchestrates the parallel-analysis phase on top of this playbook.

0
installs
▌
Long Horizon Jobs · joemccann
Durability contract for long-running agent jobs (harvesters, scrapers, ingest, audits). Use when building or running any job that processes many items over a long horizon, when a job must survive kills/context limits, or when resuming a previously interrupted harvester. Covers state.json + findings.jsonl + SUMMARY.md checkpointing, the memory-observer, and the 10-tool-call checkpoint hook.

0
installs
▌
Reliability Weekend · joemccann bundle
Weekend reliability loop - daily delta-audit of everything merged since the last audited SHA (new findings appended to RELIABILITY_AUDIT.md), then red/green remediation of EVERY verified finding on the dated PR branch, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/reliability_weekend.sh, one daily cycle at 00:00 local that runs audit, remediate, then deliver; invoke as /reliability-weekend audit, /reliability-weekend remediate or /reliability-weekend deliver.

0
installs
▌
Documentation Nightly · joemccann bundle
Nightly documentation maintainer - daily audit that classifies the documentation impact of everything merged since the last audited SHA (rolling issue labeled documentation-nightly), then smallest source-backed remediation of EVERY verified P0/P1/P2 finding on the dated PR branch documentation/<date> without inventing prose, duplicating machine truth, or touching live systems, then a deliver phase that pushes, opens one PR, gets CI green and tells the operator what to merge. Runs unattended on the always-on runner via scripts/documentation_nightly.sh, one daily cycle at 00:30 local that runs audit, remediate, then deliver; invoke as /documentation-nightly audit, /documentation-nightly remediate or /documentation-nightly deliver.

0
installs

36 published skills