By slapglif
Mathematical physics tooling suite - symbolic math, numerical physics, ML, theorem proving, verification, bioinformatics, and discrete mathematics
Bioinformatics analysis (Biopython) - sequence, protein, and structure analysis
Combinatorics and graph theory (NetworkX) - graph algorithms, enumeration, optimization
Physics ML - neural operators, PINNs, equivariant networks
Numerical physics - quantum chemistry, quantum circuits, simulations
Theorem proving - Lean 4 with RobustLeanProver (auto fallback, parallel search, caching)
Molecular biology and bioinformatics research agent. Use this agent when: (1) Analyzing biological sequences (DNA, RNA, protein) (2) Investigating protein structure and binding sites (3) Finding mutations and variations in genomes (4) Preparing computational drug discovery workflows (5) Molecular dynamics and protein-ligand interactions This agent combines computational biology tools with Theory2's physics ML capabilities.
Discrete mathematics and combinatorial optimization specialist. Use this agent when: (1) Solving graph theory problems (coloring, matching, flows) (2) Tackling NP-hard optimization (TSP, SAT, scheduling) (3) Enumerating combinatorial structures (partitions, permutations, graphs) (4) Network analysis and algorithm design (5) Constraint satisfaction and integer programming This agent combines classical algorithms with modern SMT solvers and heuristics.
Autonomous mathematical physics problem solver. Use this agent when the user needs to: (1) Solve physics problems requiring multiple computational steps (2) Explore mathematical structures (Lie algebras, group theory) (3) Perform quantum chemistry or circuit calculations (4) Train physics ML models (FNO, E3NN, PINNs) (5) Prove mathematical theorems This agent has access to the full Theory2 tooling suite.
Cross-validation and verification agent for physics calculations. Use this agent when: (1) Results need independent verification (2) Comparing theoretical predictions with experimental values (3) Checking consistency across different computational methods (4) Validating new physics calculations before publication This agent specializes in multi-method verification and discrepancy analysis.
Automated theorem proving agent using Lean 4 with RobustLeanProver. Use this agent when: (1) User needs to prove a mathematical theorem (2) Verifying mathematical claims formally (3) Building proof chains for complex theorems (4) Exploring proof strategies for hard problems This agent uses intelligent tactic selection, parallel search, and proof caching.
Admin access level
Server config contains admin-level keywords
Uses power tools
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.
Uses Bash, Write, or Edit tools
Uses Bash, Write, or Edit tools
A comprehensive mathematical physics tooling plugin that provides access to symbolic math, numerical physics, physics ML, theorem proving, and scientific verification through the Theory2 CLI.
/home/mikeb/theory2# Clone or copy to Claude plugins directory
cp -r theory2-plugin ~/.claude/plugins/
# Or use with --plugin-dir
claude --plugin-dir /path/to/theory2-plugin
/theory2-physics:symbolic compute-e7-alpha --verify
/theory2-physics:numerical quantum-chemistry --molecule="H2O" --method=dft
/theory2-physics:ml solve-pde --pde-type=heat --iterations=10000
/theory2-physics:prove lean --statement="2 + 2 = 4" --tactic=norm_num
/theory2-physics:verify cross-check --claim="alpha_inv=137"
The plugin provides rich MCP tools with detailed descriptions:
theory2_symbolic_compute_e7_alpha - E7 Lie algebra → fine-structure constanttheory2_symbolic_lie_algebra - Query Lie algebra propertiestheory2_symbolic_eval/simplify/solve/diff/integrate - SymPy operationstheory2_numerical_quantum_chemistry - HF/DFT/CCSD calculationstheory2_numerical_quantum_circuit - Quantum circuit simulationtheory2_ml_train_fno/train_e3nn/solve_pde - Physics ML modelstheory2_prove_lean/search - Lean 4 theorem provingtheory2_verify_cross_check - Multi-method verificationThe plugin implements a rigorous validation workflow:
theory2-plugin/
├── .claude-plugin/
│ └── plugin.json # Plugin manifest
├── commands/ # Slash commands
│ ├── symbolic.md
│ ├── numerical.md
│ ├── ml.md
│ ├── prove.md
│ └── verify.md
├── agents/ # Autonomous agents
│ ├── physics-solver.md
│ └── physics-verifier.md
├── skills/
│ └── theory2-physics/
│ └── SKILL.md # Usage skill
├── hooks/
│ └── hooks.json # Event hooks
├── mcp/
│ └── theory2_server.py # MCP server
└── .mcp.json # MCP configuration
MIT
MCP server for Maestro mobile testing + ADB toolkit - Playwright-style control for iOS/Android apps
End-to-end ML training ecosystem powered by SkyPilot. Launch, monitor, fix, iterate, ablate, and serve models on any cloud GPU with production-grade frameworks.
Zero-config P2P networking — connect agents to a swarm for messaging, file transfer, and tunneling
Multi-agent message bus for coordinating AI agents across sessions via pub/sub, request/response, and broadcast patterns
npx claudepluginhub slapglif/theory2-physics-plugin --plugin theory2-physicsThree AI models, one synthesis — multi-model research workflow for scientific domains
Multi-agent collaborative theorem proving with LeanTree and Ensue Memory Network
Life sciences computational skills for scientific AI agents — 197 skills covering genomics, proteomics, drug discovery, biostatistics, scientific computing, and scientific writing
Core tools for fundamental physics research: arXiv access, paper management, and literature workflows
Solve competition math (IMO, Putnam, USAMO) with adversarial verification that catches what self-verification misses. Fresh-context verifiers attack proofs with specific failure patterns. Calibrated abstention over bluffing.
Autonomous research loops with 10 commands. Generalizes Karpathy's autoresearch loop to any domain with mechanical evaluation, overnight persistence, and zero dependencies.