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 instructions/liustack/pptwise/agents-mdgit clone --depth 1 https://github.com/liustack/pptwiseWhat 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 | $0.00897 | $0.00897 |
| Opus 5 | $0.00449 | $0.00449 |
| Sonnet 5 | $0.00179 | $0.00179 |
| Haiku 4.5 | $0.00090 | $0.00090 |
Grade A, and why
pptwise AGENTS.md 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 2d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Overview for AI Agents
Goal
pptwise turns semantic IR into native editable PPTX. It ships as a skill folder under skills/pptwise/ and as a DSH plugin through dsh/index.js plus cordis.patch.yml. The stable public surface is the CLI, IR schema, theme schema, deck project format, skill, and DSH plugin. JavaScript internals carry no semver promise. See docs/internal-api.md.
Authoring model
The causal chain is:
intent -> narrative -> theme -> spec -> fill -> render
The four public nouns are:
- theme: one complete self-contained v2 file with style, optional brand configuration, occasion metadata, identity strength, and a menu. The 24 built-ins are factory presets for copying.
- spec: a version 1 semantic contract that binds one theme and locks narrative, page order, id, type, heading, and content kind.
- component: one of 37 typed units that fill a page.
- kind: one of 11 content-page moves. The vocabulary is
points,list,comparison,process,data,photo,statement,quote,fact,evidence, andhierarchy.
Boundary pages are cover, chapter, and ending. Content pages require kind. The bound theme menu maps each boundary type and each offered content kind to exactly one internal face. Authors never write face ids. IR v5 has no seed, layout, beat, or arrangement.
See docs/concepts.md, docs/menu-lookup.md, docs/themes.md, docs/ir.md, and docs/deck-projects.md.
Architecture
IR and project content flow through one shared theme-menu route into the React SVG page system under src/layouts, src/components, src/motifs, and src/render. Checks live under src/audit. Static SVG markup goes through src/pptx, svg2pptx, PptxGenJS, and JSZip patches before producing .pptx.
Validation, capacity, asset briefs, and rendering consume the same resolved menu route. Pacing supplies the editorial budget and body baseline. The chosen face supplies physical capacity. Pacing never chooses a face.
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.
- 2d ago First seen · 57 lines · 897 tokens per session scan A dfd1994be3e4
pptwise AGENTS.md is an instructions file published in the GitHub repository liustack/pptwise (10 stars, last pushed 2d ago), licensed MIT. It adds 897 tokens to every session, about $0.0045 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 2026-08-31.
Other instructions, from other repositories
dsh-worktable AGENTS.md
Instructions for Aisland-SJL/dsh-worktable, covering dsh-worktable 项目规则, 协作方式（用户定案，最高优先级）, 边界, 构建与验证 and 领域约定（会话中必须遵守）.
tingly-box CLAUDE.md
Instructions for tingly-dev/tingly-box, covering tingly-box, guideline, ux-first (highest priority), frontend and icons.
awesome-deepseek-harness-plugins AGENTS.md
Instructions for imsai-sh/awesome-deepseek-harness-plugins, covering repository instructions, the submission gate is the product, generated files, cross-repo contracts (no ci spans both repositories) and permanent urls.
dsh-config-manager AGENTS.md
Instructions for xiajiajun516/dsh-config-manager, covering agents.md — dsh config manager 仓库协作指南, 🗣️ 语言与交流, 📦 项目概览, 🗂️ 仓库结构与职责 and ui 分层（改动前必读，避免放错层）.
odai AGENTS.md
Instructions for orziz/odai, covering 仓库维护约束, odai 必用, 官方 skills 单一事实源, 版本发布约束 and dsh 集成修改边界.
dsh-mcp-panel AGENTS.md
Instructions for PerryLink/dsh-mcp-panel, covering agents.md, layout, hard rules applied here, config and build.