By han-xudong
Source-grounded engineering philosophy skills for Claude Code
Summarize the personas in this plugin for the request in `$ARGUMENTS`.
Read `docs/personas.md` and the most relevant skill files in this plugin. Recommend one primary persona, or at most one complementary pair, for the task described in `$ARGUMENTS`.
Behavioral guidance derived from Andrej Karpathy's public writing and talks on agentic coding, task framing, scope control, and explicit success criteria.
Behavioral guidance derived from Charity Majors' public writing on feedback loops, production validation, operational pragmatism, and observability.
Behavioral guidance derived from Donald Knuth's public work on craftsmanship, invariants, explanation quality, and disciplined optimization.
Behavioral guidance derived from John Carmack's archived .plan files and public technical talks on direct implementation, cost visibility, and measurement-first engineering.
Behavioral guidance derived from John Ousterhout's public work on software design, complexity management, deep modules, interface depth, and comments.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Source-grounded engineering personas for coding agents.
This repo turns public engineering philosophies into usable skills: not imitation, not quotes, and not fan-fiction. Each persona is reduced to operational guidance an agent can actually use under pressure: principles, behavioral constraints, and a default execution bias.
Install · Use with Cursor · Examples · Included Personas
Most persona prompting is shallow. It tends to be biographical, theatrical, or vague enough that it does not change engineering behavior in a reliable way.
This project is trying to do the opposite:
If a claim cannot be defended from public material, it should not be here.
This is the shape of prompt the repo is meant to make possible:
Use `andrej-karpathy`.
Task: Add email validation to the signup form.
Before writing code, surface any ambiguity in the request, define one falsifiable success check, and make the first change as small and testable as possible.
And when a hybrid is actually justified:
Use `kent-beck` + `john-ousterhout`.
Take small safe steps with fast feedback, but keep pushing complexity behind a deeper module boundary instead of leaking bookkeeping to callers.
That is the product: sharper operating bias, not stylistic mimicry.
The personas are grouped by the kind of engineering risk they are best at controlling.
andrej-karpathy — ambiguity control, task framing, scope disciplinejohn-carmack — direct implementation, runtime cost visibility, measurementkent-beck — tiny safe steps, feedback loops, reversible changedonald-knuth — correctness, invariants, precise reasoningrich-hickey — conceptual simplicity, de-complecting, value-oriented designjohn-ousterhout — deep modules, interface quality, complexity managementsandi-metz — delaying abstraction, truthful code under changerob-pike — regularity, clear flow, low-ceremony toolsjose-valim — explicit boundaries, composable systems, runtime-aware toolingcharity-majors — observability, production feedback, rollout realismPick the persona that matches the dominant risk in the task.
npx claudepluginhub han-xudong/elite-principles-skills --plugin elite-principles-skillsComprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Write feature specs, plan roadmaps, and synthesize user research faster. Keep stakeholders updated and stay ahead of the competitive landscape.
Curate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.