HarleyCoops

HarleyCoops/Math-To-Manim

Instructions file CodexOpenCode

Instructions for HarleyCoops/Math-To-Manim, covering agent guide, what this repo is, layout, working rules and quick verification.

HarleyCoops/Math-To-Manim

Instructions file

Instructions for HarleyCoops/Math-To-Manim: Rankings, higher = better. Cost reflects what I actually pay (OpenAI has really generous limits), not list price. Intelligence is how hard a problem you can hand the model unsupervised. Taste covers UI/UX, code quality, API design, and copy.

HarleyCoops/Math-To-Manim

Skill Claude CodeCodex

Operate Math-To-Manim with Hermes: inspect artifacts, run deterministic checks, use Codex-backed codegen, render/review Manim outputs, and avoid committing generated media.

HarleyCoops/Math-To-Manim

Agent

Stage 2 of the GLM chain. Builds the reverse knowledge tree. Depth 0 is the target claim. Edges are prerequisites. The spine starts at assumed foundations and walks forward to the claim. Cite canonical sources only when naming them.

HarleyCoops/Math-To-Manim

Agent

Stage 5 of the GLM chain. Writes the shot list against the curriculum and dossier. Camera grammar is setcameraorientation / movecamera only. Reference images are visual seeds, never teaching sources.

HarleyCoops/Math-To-Manim

Agent

Stage 6 of the GLM chain. Writes the ThreeDScene contract and the full Manim source as glmscene.py, then calls local verifyscene until it passes. Camera moves use setcameraorientation or movecamera only.

HarleyCoops/Math-To-Manim

Agent

Stage 3 of the GLM chain. First forward pass. Turns the reverse spine into acts, one new idea each, joined only by curiosity. No tools.

HarleyCoops/Math-To-Manim

Agent

Stage 1 of the GLM chain. Distills a prompt or photographed homework page into audience, core claim, scope, and the one big zoom. Vision via chat/completions image parts when attached. Writes no code.

HarleyCoops/Math-To-Manim

Agent

Stage 4 of the GLM chain. Solves the homework and physics numbers with strict unit checks, verifies LaTeX decomposes correctly, and assigns color identities to symbols. Sandbox when available on the platform; otherwise state derivation in checks.

HarleyCoops/Math-To-Manim

Agent

Stage 2 of the Grok chain. Builds the reverse knowledge tree. Depth 0 is the target claim. Edges are prerequisites. The spine starts at assumed foundations and walks forward to the claim. Optional websearch only for canonical names.

HarleyCoops/Math-To-Manim

Agent

Stage 5 of the Grok chain. Writes the shot list and generates 1 to 3 art direction stills. X search is only a visual seed, never a teaching source.

HarleyCoops/Math-To-Manim

Agent

Stage 6 of the Grok chain. Writes the ThreeDScene contract and the Manim source, then calls local verifyscene. Camera moves use movecamera or setcameraorientation only.

HarleyCoops/Math-To-Manim

Agent

Stage 3 of the Grok chain. First forward pass. Turns the reverse spine into acts, one new idea each, joined only by curiosity. No tools.

HarleyCoops/Math-To-Manim

Agent

Stage 1 of the Grok chain. Distills a prompt or photographed homework page into audience, core claim, scope, and the one big zoom. Uses vision when an image is attached. Writes no code.

HarleyCoops/Math-To-Manim

Agent

Stage 4 of the Grok chain. Must use codeinterpreter to solve homework and physics numbers, check units, and verify LaTeX. websearch only for cited constants.

HarleyCoops/Math-To-Manim

Skill Claude CodeCodex

This skill should be used when the user asks to "create a math animation", "animate a mathematical concept", "generate Manim code", "visualize [topic] with animation", "explain [concept] visually", "create an educational video", "build a Manim scene", or mentions "reverse knowledge tree", "prerequisite discovery", or…

HarleyCoops/Math-To-Manim

Instructions file GitHub Copilot

Instructions for HarleyCoops/Math-To-Manim, covering github copilot instructions for math-to-manim, project overview, key technologies, development setup and installation.

HarleyCoops/Math-To-Manim

Instructions file GitHub Copilot

Guidelines for working with AI agents in the Math-To-Manim pipeline.

HarleyCoops/Math-To-Manim

Instructions file GitHub Copilot

Guidelines for creating and modifying Manim animation examples.

HarleyCoops/Math-To-Manim

Instructions file GitHub Copilot

Guidelines for writing and maintaining tests in Math-To-Manim.

HarleyCoops/Math-To-Manim

Agent

Stage 2 of the Mythos chain. Builds the reverse knowledge tree — walks backward from the target concept to first principles, mapping every prerequisite the film must teach or assume.

HarleyCoops/Math-To-Manim

Agent

Stage 5 of the Mythos chain. Converts acts plus mathematics into a beat-by-beat shot list using the Mythos camera grammar — headlines, zooms into terms, pull-backs, 3D set pieces, captions.

HarleyCoops/Math-To-Manim

Agent

Stage 3 of the Mythos chain. Converts the knowledge map's spine into a dramatic act structure — what is taught when, what question pulls the viewer into each act.

HarleyCoops/Math-To-Manim

Agent

Stage 1 of the Mythos chain. Distills a raw user prompt into a cinematic intent brief — audience, core claim, emotional arc, scope. Use first when turning any math/physics topic into a film.