modern-web-guidance
claudeKeep your coding agent up to date with the latest web best practices
P2P — Prompt to Prompt, Claude Edition: a meta-prompt plugin that writes task-specific prompts for any model.
Plugin p2p from marketplace P2P in sanic732/P2P. Commands read from the repo's manifest — 30 days ago.
capsule/p2p:capsule — save or load session context via CAPSULE protocol.
chain/p2p:chain — run custom agent chain in specified order.
chart/p2p:chart — show or update CHART task map.
explore/p2p:explore — activate Exploration Mode (Cortex Patch A).
karpathyP2P Karpathy Coding Mode — precision behavioral constraints for surgical code edits. Minimalist approach — Template M + PRE_CODE_DECLARATION. Composable with agentic (M+R) and IDE (M+N) templates.
langP2P LANG — set the language P2P talks to you in: any language of the world by name or code (español, Deutsch, uk, 日本語), or auto. Call it with /p2p:lang [language|auto].
menuP2P full UI menu — every item [0]–[42], the modes and the choice window. Call it with /p2p:menu.
p2pP2P meta-prompt system for Claude: turns a raw request into a prompt contract for the target model. Call it with /p2p [task] or one of the /p2p:[command] commands.
quorumP2P QUORUM — the 8 P2P agents (architecton, axiom, datos, helios, iris, tecton, vector, anon) look at a prompt, design or problem from several sides, then one synthesis. Call it with /p2p:quorum [mode] [task]. Not for plain prompt generation (use /p2p).
ragP2P RAG — sources and retrieval for NotebookLM, Claude Projects, Cowork or any RAG backend, in four modes: prep (clean a document into RAG-ready Markdown), grounding (a prompt that answers only from the sources, per model), router (where to put the data and which scheme: naive, RAPTOR, LongRAG), not
reportP2P REPORT — an anonymised report for the author of P2P: which features ran, routes, target models, where P2P went wrong, your rating. Shown to you in full; you send it yourself. Call it with /p2p:report.
scope/p2p:scope — activate SCOPE.HELM for large multi-step tasks.
teacherP2P course: 5 levels (Quickstart → Commands → Agents → QUORUM → SCOPE.HELM) with exercises, sandbox tasks and Q&A. Call it with /p2p:teacher [level|question]. Not for generating prompts (use /p2p).
anonP2P QUORUM role ANON — code prompt specialist, optimization specialist, algorithm expert, performance analyst. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
architectonP2P QUORUM role ARCHITECTON — senior integrator, holistic view, conflict resolver. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
axiomP2P QUORUM role AXIOM — devil’s advocate, critical verifier, weak point finder. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
datosP2P QUORUM role DATOS — data analyst, fact-checker, empirical verifier. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
heliosP2P QUORUM role HELIOS — final synthesizer, executive presenter, actionable output generator. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
irisP2P QUORUM role IRIS — problem space cartographer, research specialist, unknown discovery. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
tectonP2P QUORUM role TECTON — system architect, task decomposer, structural planner. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
vectorP2P QUORUM role VECTOR — security engineer, privacy guardian, threat modeler; de-noising and veto. Use only inside a P2P QUORUM, sub-QUORUM, /p2p:chain run or a direct call from P2P menu item [8]; do not delegate other tasks to it.
Keep your coding agent up to date with the latest web best practices
Keep your coding agent up to date with the latest web best practices
2 skills
GSD Core is a meta-prompting, context engineering, and spec-driven development system for AI coding agents.
72 skills · 72 commands · 64 agents · hooks
Keep your coding agent up to date with the latest web best practices
2 skills
Author Bento presentations — self-contained .bento.html decks with morph transitions, live charts and speaker notes. Starts from nothing: downloads the latest Bento app automatically.
1 skill
Strategic foresight intelligence using IFTF methodology. Two prediction modes: Soft Predict Future (Claude-native skill, instant, claude.ai + Claude Code) and Hard Predict Future (deterministic 12-step pipeline, Python-computed — Claude Code only). Both output four independent futures (Probable/Plausible/Possible/Preferable) with per-stakeholder conditional analysis. Try: Will India become the global AI leader? Will OpenAI or Anthropic win?
2 skills
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.