A Complete faster runtime for coding agents. Optimize the journey, not the hop. Make coding agents 25% faster and 30% cheaper on average while keeping the quality same or more. Same Task, Same Quality, Faster and Cheaper.
This repository also configures its own agents. See what lemoncrow tells them →
Skill Claude Code
Measures what LemonCrow actually buys you on your repo and prompts — offline from session history (free) or online as a real A/B run against the host you are in, vanilla vs LemonCrow (real API spend). Does not benchmark LemonCrow internals — dev suite commands at the bottom for that.
Skill Claude Code
Thin pass-through to the real lc CLI (already built, already validated) — lc rt agent / lc rt skill / lc settings set / any other documented subcommand. Run the right CLI command, relay its output. Never invent cost numbers, validation, or confirmation logic here; the CLI already owns all of that. Not the cross-host…
Skill Claude Code
Run one multi-step task end-to-end on the right execution surface — a background task, a durable workflow, or a direct subagent — and hand back a runid to track. Use for 'orchestrate this', 'run this in the background', 'kick off a multi-step run', or /orchestrate.
Skill Claude Code
Verify a code change against measured performance gates — latency & tail-latency regression, profiler-confirmed hot paths, memory/leak soak, I/O & wire budgets, and scaling — by running it, not reading it. Use for 'perf review', 'did this slow down', 'check for a regression', or /perf-review. Enforces performance…
Skill Claude Code
Retrieve what LemonCrow already learned from your past sessions — semantic recall over indexed sessions, durable facts, and extracted lessons. Use for 'have we seen this before', 'what did we learn about X', 'what does LemonCrow remember', or /recall. Read-only; importing, indexing, and extraction run in the LemonCrow…
Skill Claude Code
Launch N parallel attempts at one hard task in isolated worktrees, then reduce to the best by merge, fitness, findings-union, or vote — for 'try this N ways', 'swarm on X', 'optimize/tune X', 'find all bugs', or /swarm. Use /orchestrate for a single sequential run instead.
Skill Claude Code
Verify a shipped UI against objective design gates — WCAG accessibility, design-token fidelity, responsive/render integrity, interaction & state coverage, and visual regression — by rendering it in a real browser. Use for 'ux review', 'a11y check', 'does this UI hold up', or /ux-review. Enforces design quality; does…
Skill Claude CodeCodex
Autonomous unattended mode.
Skill Claude CodeCodex
Software engineer on a lean toolset (token-heavy tools stripped): run tasks end to end.
Skill Claude CodeCodex
Software engineer: ship the asked-for change end to end — locate, edit, verify, report.
Skill Claude CodeCodex
Isolated plan execution mode.
Skill Claude CodeCodex
Read-only explorer: locate the code answering the question, cite it by stable anchor, report fast.
Skill Claude CodeCodex
Catch-all agent: work fitting no specialized role — mixed research+implementation, ad hoc investigation, multi-step chores across code and shell. Never assume every task is a code change. Pure code change → code; accepted plan → execute; checked deliverable → solve.
Skill Claude CodeCodex
Read-only task planning mode.
Skill Claude CodeCodex
External web research mode.
Skill Claude CodeCodex
Adversarial general purpose review mode.
Skill Claude CodeCodex
Autonomous focused hard problem solver mode.
Skill Claude CodeCodex
Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens"…
Skill Claude Code
Set up LemonCrow Review surfaces and self-contained execution runtimes for any repository.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: