Deonte Cooper

Set up local semantic memory search in OpenClaw using Ollama + nomic-embed-text. Free, private, offline-capable. Replaces cloud embedding APIs (OpenAI, Gemini) with a locally-running model. Trigger phrases: ollama memory, local embeddings, private memory search, offline memory, free embeddings, nomic-embed-text.

Import AI conversation exports (OpenAI ChatGPT and Anthropic Claude) into your agent's episodic memory. Parses export files, writes daily memory summaries, and deduplicates on re-runs. Trigger phrases: import chat history, ingest conversation export, import chatgpt export, import claude export, load AI conversation history.

Extract structured learnings (lessons, decisions, patterns, dead ends) from AI conversation exports using a local Ollama model or any OpenAI-compatible API. Pairs with chat-history-importer. Trigger phrases: extract learnings from conversations, analyze chat exports, mine conversation insights, extract lessons from chats, chat learnings extractor.

Minimize Anthropic Claude API costs through model selection, prompt caching, batching, and cost tracking. Trigger phrases: reduce API costs, optimize Claude spending, save on API calls, Claude cost optimization, cheaper Claude models.

Run Freqtrade backtests for cryptocurrency trading strategies, interpret results, and iterate. Use when downloading historical data, running a backtest, reading backtest output, comparing strategies, or deciding whether a strategy is ready for live trading. Trigger phrases: backtest freqtrade, run backtest, download freqtrade data, compare strategies, is my strategy good, freqtrade results.

Develop, iterate, and improve Freqtrade cryptocurrency trading strategies. Use when writing a new strategy, improving an existing one, analyzing why a strategy is losing, or understanding which indicators to use. Covers strategy anatomy, key configuration parameters, proven entry/exit patterns, and the iteration workflow. Trigger phrases: write freqtrade strategy, improve strategy, why is my strategy losing, freqtrade indicators, strategy not profitable, freqtrade entry conditions.

Shell aliases and helper commands for Freqtrade (cryptocurrency trading bot) that speed up common tasks. Use when setting up Freqtrade shortcuts, downloading market data quickly, running backtests from the command line, or controlling the bot. Covers bash/zsh (Linux/macOS) and PowerShell/Command Prompt (Windows). Trigger phrases: freqtrade aliases, freqtrade shortcuts, ftdata, ftback, freqtrade commands, freqtrade windows.

Complete setup guide for running Freqtrade (cryptocurrency trading bot) legally in the United States. Use when setting up Freqtrade for the first time, choosing a US-compatible exchange, configuring API keys securely, or troubleshooting US-specific exchange issues. Covers Kraken setup, Docker configuration, API key security, and dry-run testing. Trigger phrases: freqtrade US, US exchanges freqtrade, freqtrade legal US, freqtrade Kraken setup, freqtrade docker config.

Write clear, structured GitHub pull request descriptions using a standardized template. Use when creating a PR, updating a PR description, or when asked to document what a branch changes. Automatically fills in summary, type of change, description of changes, impact, and testing notes based on the diff or context provided. Triggers on phrases like "create a PR", "write the PR description", "open a pull request", "update the PR body", "document this PR", or when a branch is ready to merge.

Run two OpenClaw agents simultaneously — a paid Anthropic agent and a free OpenRouter agent, each with its own Telegram bot. Trigger phrases: multi-agent setup, add a second agent, free agent openclaw, run two agents, openrouter openclaw, parallel agents, cost optimization agent.

Autonomously fix failing CI on Rails PRs using a tiered escalation loop. Use this skill whenever a Rails pull request has failing CI — RSpec failures, RuboCop offenses, migration errors, factory issues, seed data problems, or build environment failures (yarn, npm, Tailwind, missing system deps). Handles the full cycle without human intervention: pull logs, fix with a fast model, escalate to a stronger model if needed, notify human when green or stuck. Never merges — human always merges. Trigger phrases: fix CI, CI is failing, CI is red, watch the PR, fix the tests, the build is broken.

RSpec testing standards and best practices for Rails applications. Use when writing new tests, reviewing test quality, debugging factory errors, setting up FactoryBot, or enforcing single-expectation patterns. Also use when a test fails due to factory misconfiguration, wrong association keys, or missing role traits. Triggers on phrases like "write a test", "add specs", "factory error", "test is failing", "how should I test this", or when reviewing test code in a Rails project.

Tailwind CSS component patterns for Ruby on Rails ERB views. Use when building UI components in Rails, creating shared partials, implementing dark mode, writing badge/button/card/table components, or setting up a consistent design system. Trigger phrases: tailwind rails, rails UI components, ERB tailwind, rails dark mode, tailwind badge, tailwind button rails, shared partials tailwind.

Tiered model selection and cost optimization for multi-agent AI workflows. Use this skill whenever you are choosing a model for a task, spinning up a sub-agent, setting up cron jobs or heartbeats, or trying to reduce API spend. Also use when the user says "save costs", "which model should I use", "optimize model usage", "this is getting expensive", or when delegating any task to a sub-agent. Works with any AI provider.

Create Twitch clips of the current live stream via the Twitch API. Use when a streamer says "clip that", "clip the last X seconds", "make a clip", "clip it", or any variation requesting a clip of the current broadcast. Requires Twitch API credentials configured in environment variables. Has a 30-second cooldown between clips to prevent spam.