By Parijat-18
Compile a research paper and its citation neighborhood into an implementation-ready memory for Claude Code. Domain-neutral (ML, physics, chemistry, biology). MCP-first; hard policy enforcement; nested sub-skill architecture; cross-session memory via decisions.md + sessions/.
Cross-check the repo against the compiled paper-compiler context. Domain-neutral — works on ML reproductions, physics simulations, chemistry pipelines, biology protocols. Dispatches per-category audit sub-skills (audit-method / audit-objective / audit-data / audit-procedure / audit-evaluation / audit-baseline / audit-theory). Warn-only — never auto-fixes.
Use when the user asks to "build research context", "compile this paper", "ingest <paper>", "refresh research/", or hands you a fresh arXiv / DOI / S2 id / open-access URL and wants the full paper-compiler pipeline run. Works for any scientific or engineering domain — ML, physics, chemistry, biology, economics, climate, etc.; categories and roles are domain-neutral. Starts the CLI compile as a background process with real-time log monitoring so the user can track progress. Manual-only; Claude never auto-invokes this.
Compare two compiled `research/` artifacts. Use when the user says "compare paper A with paper B", "how does paper X's method differ from paper Y's", or wants a meta-analysis across two papers they've already compiled. Emits `comparison-report.md` with cross-paper atom alignment, conflicting `contradicts` edges, and missing atoms in either side.
Resume from a prior session. Use when the user says "continue from yesterday", "what was I doing", "pick up where we left off", or returns to a paper repo after time away. Top-level entry point — doesn't assume the user is mid-implementation (use `/paper-compiler:use-research-context continue` for that). Surfaces a one-screen summary of recent sessions and decisions, then dispatches.
Implementation work derived from a compiled paper. Domain-neutral — ML, physics, chemistry, biology, economics, climate, etc. Routes to a category-specific sub-skill (implement-method / implement-objective / implement-data / implement-procedure / implement-evaluation / implement-baseline / debug-divergence) via `scripts/select-playbook.sh`.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
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.
A research paper is a compressed implementation artifact. The detail a coding agent needs to actually reproduce it lives across the citation neighborhood — in the cited methods, datasets, baselines, prior architectures, evaluation protocols. research-compiler is a Claude Code plugin that compiles that neighborhood into a queryable Graph RAG store and a Karpathy-style llm-wiki, then teaches Claude — through skills and MCP tools — which lever to pull for which sub-task.
A single command turns one arXiv ID into ~430 papers, ~165 implementation atoms, ~19,000 indexed chunks (prose, tables, captions, equations), three detected research communities, and a regenerable wiki — all sitting in research/ next to your code, addressable by stable UIDs, served read-only over 26 MCP tools.
PaperBench (2025) measured what we already suspected: frontier models reproduce papers at ~21%, and the dominant failure mode is missing implementation context — the kind of detail that lives one citation hop away. Existing paper-to-code systems start from the target PDF and ignore that neighborhood. We compile it.
The thesis being tested, in three operational conditions:
| Condition | Setup | Predicted outcome |
|---|---|---|
| A | Claude Code + target paper PDF | baseline |
| B | Claude Code + compiled research.md brief | most of the lift |
| C | Claude Code + brief + Graph RAG MCP + wiki | +10pp over A, hallucination halved, atom coverage ≥1.5×, last 3-5pp of accuracy on cross-paper queries |
The plugin succeeds or fails as that research claim. The evaluation rubric lives in docs/05-evaluation-plan.md.
Three primitives — skills, MCP server, forked subagent — composed into a single plugin so the workflow is one prompt away.
┌────────────────────────────────────────────────────────────────────┐
│ Claude Code session (user-facing) │
│ /paper-compiler:build-research-context arxiv:2603.19312 │
│ /paper-compiler:use-research-context (auto-invoke) │
│ /paper-compiler:audit-against-research (auto-invoke) │
│ /paper-compiler:wiki-query / wiki-ingest / wiki-lint │
│ 15 mcp__paper-compiler__* tools │
└─────────┬──────────────────────────────────────────▲───────────────┘
│ queries │ structured evidence
▼ │
┌────────────────────────────────────────────────────────────────────┐
│ paper-compiler MCP server (read-only) │
│ sqlite + sqlite-vec + FTS5; lazy-loaded; ~72 MB per paper │
└─────────┬──────────────────────────────────────────────────────────┘
│
▼
┌────────────────────────────────────────────────────────────────────┐
│ research/ (lives in your repo, git-friendly) │
│ research.md — ≤ 8000-token brief │
│ research.db — Graph RAG store │
│ SCHEMA.md — DB schema reference for Claude │
│ evidence/<atom>.md — per-atom verbatim spans │
│ wiki/ — Karpathy llm-wiki (atoms, papers, │
│ communities, promoted answers, log) │
└─────────▲──────────────────────────────────────────────────────────┘
│ writes (compile-time only)
│
┌────────────────────────────────────────────────────────────────────┐
│ paper-compiler CLI — runs in background, progress monitored │
│ resolve → acquire → parse → expand → classify → atom-extract → │
│ score → render → build DB → communities → wiki │
└────────────────────────────────────────────────────────────────────┘
The strict separation is the point. The CLI never serves runtime queries; the MCP server never writes. A stale DB diagnosed without re-running a compile. A buggy tool replaced without re-acquiring papers. The wiki regenerated as a pure function of the DB.
npx claudepluginhub parijat-18/research-compiler --plugin paper-compilerPermanent coding companion for Claude Code — survives any update. MCP-based terminal pet with ASCII art, stats, reactions, and personality.
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.
Intelligent prompt optimization: injects the right context at the right moment so Claude lands a better first output. Clarifies vague prompts with research-based questions, plus targeted nudges for approach selection, plan readability, workflow routing, background execution, subagent routing, output readability, user-decision questions, and plan-mode assessment
Complete 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.
Memory compression system for Claude Code - persist context across sessions
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.