caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Physics research around a research.md backbone: one research cycle, investigation, verification, writing, learning, brainstorming and method distillation.
Plugin aitp from marketplace aitp-protocol in bhjia-phys/AITP-Research-Protocol. Commands read from the repo's manifest — yesterday.
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.
aitp-distillTurn demonstrated research procedures into focused Skills, or revise them from use. Apply when work reveals a transferable non-obvious choice, diagnostic sequence, failure-prevention method or correction to an existing procedure, or when explicitly requested. Concepts and conclusions stay in notes.
aitp-human-brainstormingDevelop physics research questions and consequential choices with the researcher. Use for explicit brainstorming or unresolved choices of algorithm, research route, branch scope, assumptions or success criteria. Reuse settled decisions and continue routine work without repeated questions.
aitp-human-learningGuide interactive physics learning and respond to conceptual difficulties. Adapt to the learner's prerequisites and intended depth, explain a coherent step, and use targeted feedback when useful. Deliver requested complete explanations without compulsory quizzes or unsupported claims of mastery.
aitp-memoryStart AITP research work and substantive physics discussions here. Runs the research cycle around research.md, the backbone at every level (root, topic and branch) - recover the context, decide what the task settles, use the skills it needs, keep outcomes in their homes and reachable, and close with
aitp-researchInvestigate theoretical and computational physics - derivations, synthesis across sources, discriminating calculations, implementation, debugging and benchmarks. Runs inside aitp-memory's research cycle; agree on independent branches and material changes of objective or route, then continue within t
developing-librpaReview, simplify, or extend LibRPA numerical code by tracing its active GW/RPA paths, API and driver boundaries, matrix ownership, and relevant regressions. Use for LibRPA source development and PR review, not merely to launch an existing calculation.
ren-gw-notationWrite and check detailed LRI/LibRI, GW and RPA derivations for papers and notes using notation grounded in Xinguo Ren's coauthored work. Derive endpoint tensors, all four response and self-energy contractions, periodic band vertices, screening and quasiparticle equations with explicit indices, signs
ren-gw-writingDraft, organize or revise GW, RPA, RI/NAO and LibRPA papers and research or teaching notes using argument and section-level decisions studied in Xinguo Ren's coauthored papers. Use for outlines, introductions, theory-to-algorithm exposition, results, captions and conclusions; coordinate with ren-gw-
aitp-verifyCheck research claims, results, numbers, references and documents, or assess another model's review. Use before a new, changed or disputed consequential claim enters a decision, a main note's opening or a manuscript, or when asked to check, review or audit. Reuse applicable checks for unchanged clai
aitp-writingWrite and revise physics research notes, expand lecture derivations for a learner, and develop clear arguments with linked evidence. Runs inside aitp-memory's research cycle. Produce or revise a LaTeX manuscript or PDF only when requested.
aitp-reviewerIndependent read-only check that an AITP research record is faithful to the work - complete, sound, preserved and coherent. Use when the record must stand without the conversation (handoff, session end, delegation, after a context compaction), before something outside the conversation relies on a re
cat "${CLAUDE_PLUGIN_ROOT}/hooks/session-start.txt"
cat "${CLAUDE_PLUGIN_ROOT}/hooks/post-compact.txt"
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
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.