eprouveze
UserA collection of practical Claude Code skills — multi-LLM evaluation, domain management, planning, writing quality, and pair-session patterns. MIT licensed.
Categories
Indexed Skills (16)
battle
Benchmark AI model combinations on the same coding task. Runs Claude, Antigravity CLI (agy — replaces the sunset Gemini CLI on consumer plans as of 2026-06-18), Codex CLI, Kimi (Moonshot API), and DeepSeek (OpenAI-compatible API) solo and in pairs, then scores each on tokens, cost, wall time, and output quality. Produces a leaderboard to inform future model selection. Use when asked to "battle", "benchmark models", "compare models", "which model is best", "pair programming battle", or "/battle".
brief
Project-aware session briefing. Loads project-intel, decisions log, deferred actions, and known-issues files before any work, so Claude starts with project context instead of cold. Use at session start, before any task, or when the user says "/brief", "brief me", "what should I know", "catch me up".
codex-write
Delegate heavy code generation to Codex CLI while Claude orchestrates, reviews, and surgically fixes. Use for any task involving 50+ lines of new code, boilerplate generation, test writing, or large refactors. Claude acts as lead engineer; Codex acts as developer. Multi-LLM aware — checks `.claude/llm-mode.json` before delegating and falls back to direct write if Codex is unavailable.
council
5-model strategic advisory. Ask the same question to five different LLM seats in parallel, save each response to a persistent file, then synthesize a recommendation. Each seat is locked to a role (Theorist, Validator, Provocateur, Diagnostician, Field Engineer) so the responses disagree by design instead of converging by accident. Use when the user says "ask the council", "get opinions", "council on", "what do they all think", "multi-model opinion", or wants strategic input from multiple LLMs before making a decision.
deslop
Remove AI writing patterns from content. Use when reviewing blog posts, marketing copy, emails, or any customer-facing text. Checks for 22 AI-tell patterns across phrasing, rhythm, and authenticity. Trigger phrases — "deslop this", "clean up AI writing", "check for AI patterns", "remove AI tells", "content quality check", "is this sloppy".
evaluate-plan
Evaluate an implementation plan against its source requirements (PRD, spec, or user request) to identify gaps, dropped features, and lost intent. Use when the user says "evaluate plan", "verify plan", "check the plan", "what did it miss", "plan coverage", or "validate plan".
mode
Toggle between single-LLM (Claude-only) and multi-LLM (delegate to Gemini, Codex, or other CLIs) modes. Shows current mode, available external CLIs, delegation stats, and logs. Use when the user says "switch to multi", "use Gemini", "delegation status", "how many delegations", or "token savings". Subcommands: `/mode` (status), `/mode multi`, `/mode single`, `/mode stats`, `/mode reset`.
model-scan
Scan LLM provider APIs for available models, fetch official docs for specs (parameters, context windows, capabilities, pricing), save timestamped doc snapshots, and optionally update a project's `CLAUDE.md` model table. Use when the user asks about current models or wants to refresh the model list — "scan models", "refresh models", "what models are available", "update model table", "model pricing", "check model IDs".
namecheap
Namecheap domain management via the Namecheap XML API. Check domain availability, register/transfer/renew domains, manage DNS records (A, AAAA, CNAME, MX, TXT, etc.), list domains, set custom nameservers, manage email forwarding, and check pricing. Use when the user says "namecheap", "check domain", "register domain", "DNS records", "nameservers", "domain transfer", "renew domain", or "domain pricing".
pair-session
AI pair programming with Claude (builder) and a second model (advisor). The human observes and can intervene. Three styles: standard (Claude builds → advisor reviews), ping-pong (TDD-style alternation), and strong-style (advisor proposes → Claude implements). Use when asked for "pair session", "claude codex pair", "ai pair programming", or "/pair-session".
second-opinion
Get a second opinion from OpenAI's Codex CLI. Default model is gpt-5.5. Three modes: review (pass/fail code review on the current diff), challenge (adversarial pass — try to break it), consult (multi-turn Q&A with session persistence). Uses ChatGPT subscription auth, not an API key. Use when asked for "second opinion", "codex review", "adversarial review", "challenge this code", "cross-model review", or "/second-opinion".
crawl
Fetch web pages that may be JS-rendered or bot-protected, returning clean markdown or HTML. Works zero-setup with a direct HTTP fetch; if you supply your own Cloudflare Browser Rendering credentials it uses a managed headless browser that renders JS and bypasses most WAFs. Use when a plain fetch returns 403, when a page is a JS-rendered SPA, or when you need reliable markdown extraction from a URL. Triggers on 'crawl this page', 'fetch this URL', 'scrape this site', 'get the content from this page', 'this page is blocked', or when a normal fetch fails on a URL.
keyword-research
Check Google Trends for keyword interest and SEO prioritization. Use when the user mentions "keyword research", "SEO", "search trends", "content strategy", "blog prioritization", "compare keywords", "trending topics", or wants to evaluate keyword opportunities.
plan
Research-first planning for any non-trivial task. Spawns parallel sub-agents to explore the codebase, checks past solutions and decisions, asks clarifying questions, then produces a structured plan stored in docs/plans/. Use when starting any non-trivial task: "plan this feature", "how should we approach...", "let's think through...", or before any multi-file change. Do NOT use for trivial single-file edits or questions that need no planning.
salesforce-reports
Create, clone, read, run, and delete Salesforce Reports from the command line via the Analytics REST API, driven by the Salesforce CLI (`sf`). Org-agnostic — works against any org you've authed with `sf`. Includes a `--setup` flow (installs the sf CLI if missing, saves a default org/API version, and an optional Global Company filter for GAM roles). Use when the user says "create a Salesforce report", "clone a report", "generate a report via sf CLI", "list/run/delete reports", "report API", or "Analytics REST reports".
update-machine
Clean up, update, and upgrade this machine's package managers — Homebrew (formulae + casks), npm globals, and (when present) pipx and uv tools — safely and in parallel. Surveys what's outdated first, runs the safe upgrades, guards against known traps (a tool losing its CLI entrypoint on upgrade, a GUI app's binary shadowing a same-named CLI), and HOLDS session-critical cask updates (the terminal running the session, the VPN) for explicit confirmation. Use when the user says "update packages", "upgrade the machine", "clean up homebrew/npm", "update everything", "package cleanup", or "/update-machine".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.