superpowers
claudecodexCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
LOCI predicts what AI-generated code will do — time, power, memory, and system behavior — during planning and code writing, before it runs.
Plugin loci from marketplace loci in auroralabs-loci/loci-claude. Commands read from the repo's manifest — 13 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
bug-reportForensic diagnostic report for LOCI — collects environment state, runs health checks, and writes a timestamped report when analysis fails or doesn't trigger. Use when the user says "bug report", "LOCI isn't working", "something is broken", "debug LOCI", or wants any LOCI failure investigated. Also "
contractAuthor and inspect this repository's Contract Envelope (.loci/contract.yaml) — the stack, timing, energy, memory and structural bounds that let LOCI judge a measurement rather than merely report it. Use when the user states a limit as a requirement: "set a budget", "add a bound", "cap X at N", "X mu
control-flowAnnotated control-flow graphs for compiled C/C++/Rust/Go and Python (Cython): per-function basic blocks, loops with derived trip counts, recursion and irreducible cycles, call sites, in text optimised for LLM analysis. Renders the graph and judges the two structural signals it determines, recursion
exec-traceAnalyze function execution timing and energy from compiled assembly. Use for the timing or energy of a specific function ("how long does X take", "exec trace"); a limit stated in the request is judged here, and a standing bound is authored by /loci:contract. A request that only states a limit ("cap
helpQuick-reference guide to LOCI — shows available skills, environment status, and troubleshooting for initialization and sign-in issues. Use when the user asks for help with LOCI, what LOCI can do, how to use it, which LOCI commands exist, or seems confused about LOCI setup or capabilities. Also when
initRecord how this project builds — one machine-local recipe (`.loci/build.yaml`) naming the target ISA, the compiler, the build system and the artifact — and the clock its board runs at, so every LOCI measurement compiles and times the way the project itself does. Use on /loci:init, "initialize LOCI",
loci-post-editCompare pre-edit and post-edit compiled artifacts to report execution timing % diff, energy consumption, and control-flow analysis. MANDATORY: invoke IMMEDIATELY when the `[loci] <file> was modified … You MUST invoke` reminder appears after an Edit/Write; also "analyze the change", "measure the edit
loci-preflightExecution-aware preflight analysis (timing/energy) on the functions an edit touches and the callees of any new code, using compiled artifacts, to catch problems while the design is still cheap to change. MANDATORY in plan mode (/plan; a [[PLAN]] prompt under GitHub Copilot CLI) when the user describ
memory-reportROM/RAM memory usage analysis for embedded firmware: section breakdown, top consumers, and region utilization from compiled ELF binaries. Use for "memory report", "ROM/RAM usage", "how much flash/RAM", "memory footprint", "memory map", "memory delta", "size impact"; no web/script projects. Do NOT in
setupInstall the loci CLI and verify the environment. Use when the user asks to set up, install, reinstall, or repair LOCI ("set up loci", "install loci", "loci is not installed", "fix my loci install"), or when a loci command fails because the CLI is absent. When the ask is just "doctor", only run diagn
stack-depthWorst-case stack depth analysis for embedded C/C++/Rust/Go and Python (Cython): call-graph traversal, per-function frame sizes, recursion detection, and stack budget pass/fail from compiled .o or linked ELF binaries. Use for what the stack costs: worst-case depth, stack overflow risk, frame size imp
trendsPer-function measurement history on the current branch: timing, energy, stack, and memory trends over time from LOCI analysis. Use when the user says "show trends", "optimization progress", "what changed on this branch" (in LOCI data), "how are my functions doing", or asks about performance trajecto
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/session-init.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/turn-clean.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/prompt-submit-turn.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/contract-guard.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/pre-edit-hook.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/post-edit-hook.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/post-bash-bypass.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/subagent-start.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/stats-flush.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/draft-pending-nudge.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/manifest-status-nudge.sh"'
bash -c 'exec bash "$CLAUDE_PLUGIN_ROOT/hooks/turn-clean.sh"'
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.
5 commands
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
Use Alibaba Open Code Review from local Codex.
2 skills
Claude Code skill for creating and presenting Slidev slidedecks
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.