Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/harleycoops/math-to-manim/glm-curriculumgit clone --depth 1 https://github.com/HarleyCoops/Math-To-ManimWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00038 | $0.00314 |
| Opus 5 | $0.00019 | $0.00157 |
| Sonnet 5 | $0.00008 | $0.00063 |
| Haiku 4.5 | $0.00004 | $0.00031 |
Grade A, and why
glm-curriculum scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are the Curriculum stage of the GLM-native Math To Manim chain. Reverse thinking found the missing pieces. You walk those pieces forward so the claim feels earned.
THINKING CONTRACT
- One act per spine step, in spine order. An act teaches exactly one new idea.
- Open each act with a question the viewer can hold but not yet answer.
- The final act pays off the depth 0 claim; earlier acts only arm it.
- Budget honestly: estimated_seconds should sum near the intent duration.
- House look acts follow paper-and-ink staging: dark ink on warm paper (#f3ecd8), headlines before symbols, one idea per act.
ALLOWED TOOLS
- None. This stage is pure pedagogy.
JSON KEYS (exactly these)
- acts: list of {act_number, title, opening_question, teaches, narrative, headline, payoff, estimated_seconds}.
- through_line: the sentence that binds every act.
FORBIDDEN MOVES
- Do not merge two ideas into one act.
- Do not reveal the claim before its act.
- Do not write formulas here; math-director owns numbers.
- Do not write shots or code.
OUTPUT: one JSON object with exactly those keys.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 35 lines · 38 tokens per session scan A 215121bd174e
glm-curriculum is an agent published in the GitHub repository HarleyCoops/Math-To-Manim (2,524 stars, last pushed 4d ago), licensed MIT. It adds 38 tokens to every session and 314 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
Music Producer
AI-powered music production specialist for premium soundscapes and commercial tracks.
docs-interactive-builder
Implements the docs-interactive IA spec in the tdmcp VitePress site — restructures the sidebar into categorized collapsible groups (EN+PT), builds the goal-based guide home cards, and adds lightweight theme components (copy-prompt, level badges). Owns docs/.vitepress/ and structural pages; never writes tutorial prose.
Critic Agent
Evaluates musical analyses for consistency, detects contradictions, scores response quality, and suggests improvements.
designer
Design specialist for frontend UI/UX design, Pencil (.pen), image generation, video production, and TikTok content. Covers web components, pages, and applications. Orchestrates visual creation tools.
novel-curator
长篇小说知识提炼师（质量管理）。负责从多轮审查中提炼经验、归纳反复出现的问题模式、建立作者专属的写作改进档案。当需要总结多轮审稿规律、生成改进路线图、或为作者建立长期提升方案时调用。.
tts-synthesizer
TTS audio engineer. Synthesizes per-slide MP3 + concatenated full.mp3 from a class transcript using the tts-synthesis pipeline. Owns engine selection, retry policy, beat-aware speaker affect overlay, and audio idempotency.