By yilu
Research Epistemics: Note, Observe, Reference, Model. Claim discipline and grounding for theoretical physics work in Claude, plus scientific package scaffolding (Python / Julia / C++).
Force-invoke renorm:derivation to start a derivation ledger. Optional argument: the observable target.
Force-invoke renorm:paper-analysis on a specific paper. Argument: arXiv ID, DOI, or local PDF path.
Force-invoke renorm:core to apply tagging + grounding discipline. Optional argument: a specific claim to interrogate.
Use when the user is making, asking about, or evaluating a research-level theoretical physics claim or comparison — about a specific system, mechanism, methodology, recent result, or unexplored regime. Applies to correlated systems, condensed matter, many-body physics, and adjacent topics. Skip for textbook recall (e.g., 'what is BCS pairing'), code/tooling, casual conversation, or algebra inside an active derivation.
Use when actively performing a theoretical derivation — including exploratory expansions and back-of-envelope estimates: writing equations, applying approximations (perturbation, mean-field, saddle-point, downfolding), expanding to a given order, computing a spectrum or response. Ledger weight scales with derivation complexity — short exploratory derivations get a lightweight 3-line ledger; full derivations get the 6-field form. Not for discussing existing derivations from a paper (use renorm:paper-analysis) or pure conceptual comparison (use renorm:core).
Use when initializing a new scientific computing package in Python, Julia, or C++. Picks the appropriate template directory based on user's request. Generates a working package skeleton with reproducibility conventions (RNG seeding, units in comments, type hints) consistent with the user's rules in ~/.claude/rules/.
Use when summarizing, analyzing, or extending claims from a specific paper (arXiv ID, DOI, or local PDF). Applies a quote-before-claim discipline: every assertion about the paper's content must be backed by a direct quote, equation reference, or figure reference. Not for general 'what does field X say' questions, which go to renorm:core; not for actively performing your own derivation, which goes to renorm:derivation.
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.
Research Epistemics: Note, Observe, Reference, Model.
A Claude Code plugin that imposes claim-tagging, literature-grounding, derivation, and paper-reading discipline on research-level theoretical-physics conversations. Bundles scientific-computing package scaffolding for Python / Julia / C++.
Status: v0.3.0 completes the initial discipline-skill set. Trigger language (the
descriptionfield in each SKILL.md) and this README are subject to revision based on real use; the skills themselves are stable. File issues if you find under- or over-firing.
In Claude Code, register the marketplace and install the plugin:
/plugin marketplace add yilu/renorm
/plugin install renorm@renorm
(The repo serves as both the plugin source and a single-plugin marketplace.)
For local development without installing, use the --plugin-dir flag:
claude --plugin-dir /path/to/renorm
| Skill | Purpose |
|---|---|
renorm:core | Tags every substantive physics claim (Derived / Assumed / Phenomenological / Exploratory / Training-knowledge). Grounds claims in literature when prior work matters: defaults to training-knowledge with explicit tagging, escalates to Zotero MCP or WebSearch on user request. |
renorm:derivation | Enforces a derivation-ledger discipline. Full derivations get a 6-field ledger (target / variables / what's neglected / starting equations / approximation level / falsification check); exploratory expansions get a lightweight 3-line ledger. References renorm:core's tagging vocabulary. |
renorm:paper-analysis | Enforces quote-before-claim when analyzing a specific paper: three-bucket reporting (Paper-claim / Inference / Connection), a mandatory Depends on: line for every quoted result, and explicit gap-naming when the user asks about regimes the paper doesn't cover. |
renorm:init-package | Scaffolds a new Python / Julia / C++ scientific-computing package with reproducibility conventions (RNG seeding, units in comments, type hints) baked in. |
/renorm <claim> — force-invoke renorm:core's tagging + grounding on a specific claim, or apply discipline to the next exchange./derive [observable] — start a fresh renorm:derivation ledger, optionally with the observable target as argument./paper <ref> — analyze a specific paper (arXiv ID, DOI, or local PDF path) under renorm:paper-analysis.The three discipline skills compose by reference rather than firing simultaneously:
renorm:core fires on research-level physics conversations outside active derivation or paper analysis.renorm:derivation fires when actively deriving and incorporates core's tagging vocabulary by reference.renorm:paper-analysis fires when a specific paper is the subject and incorporates core's tagging for non-paper claims that arise.Wrappers do not depend on core firing in parallel — they reference its vocabulary, so the discipline is preserved regardless of which skill activates.
MIT.
npx claudepluginhub yilu/renorm --plugin renormComplete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive startup business analysis with market sizing (TAM/SAM/SOM), financial modeling, team planning, and strategic research
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications