KiaroSamaAGENTS.mdKiaroSama/Revayat-Comic-Skill/AGENTS.md
agent skill that translates comics into Persian and produces a finished CBZ,
PDF or folder of pages. It ships as three things from one tree:
- a **skill** — `skills/revayat-comic/`, self-contained
kunalsuriAGENTS.mdkunalsuri/1000x-Agent-Skills/.agents/AGENTS.md
unmodified. Three rules make that honest. Check `LICENSE.txt` *inside the skill directory* first -- `docx`, `pdf`, `pptx` and `xlsx` are source-available under Anthropic's own terms and cannot be redistributed
luabaggAGENTS.mdluabagg/canhoto/AGENTS.md
redaction, categorize, …
migrations/ # Alembic scripts (SQLite)
parsers/ # Protocol, loader, scaffold (no bank logic)
exporters/ # pdf_summary
mcp/ # allowlist + Fast/MCP server → service only
service.py # façade for CLI + MCP
cli.py
examples/parsers/ # docs
matematicsolutionsAGENTS.mdmatematicsolutions/mcp-isap/AGENTS.md
them the tool hands back
repealed wording that looks current.
Acts served only as PDF return `isError` with `text_unavailable_use_pdf`. That is not the
text of the provision
mcarmona1064-cellAGENTS.mdmcarmona1064-cell/clawbridge-codex/container/AGENTS.md
natively, so you will see the image content directly. Always do this before responding.
- **`[pdf: ...]`** — call **Read** on the path. The Read tool extracts PDF text and renders pages natively
mrslbtAGENTS.mdmrslbt/pdf-it/AGENTS.md
working in this repo. Read this before making changes.
## What this project is
`pdf-it-mcp` is an MCP server + Claude Code skill that converts markdown into designed PDFs. v2.0.0
nburnsAGENTS.mdnburns/doc-search-mcp/AGENTS.md
search-mcp
An MCP server that indexes local documents (PDF, EPUB, HTML, TXT, MD) and exposes semantic search over them via the Model Context Protocol.
## Architecture
The indexing pipeline runs
OMCHOKSI108AGENTS.mdOMCHOKSI108/Multiagent-LangGraph-Orchestrated-LLM-Research-Automation-Platform/docs/agents.md
research plan.
---
### 2. Search Agent
Searches:
```text id="9jh641"
Web search
Academic search
PDF search
GitHub search
Company docs
Image search
```
Use APIs where possible. For Google images
OriginwiseAIAGENTS.mdOriginwiseAI/skills/AGENTS.md
code-plugin 与 bundle-plugin 不能复用同一条七产品长列表说明。Skill 应沿用官网定位 `Explain any topic with precision`,具体写清 PDF、PPT、网页链接和其他文档到 `precise, clearly structured explainer videos and interactive lessons`,中文对应“表达精准、结构清晰的解说视频和互动课程”;code-plugin
pauloribeiro16AGENTS.mdpauloribeiro16/my-skills/AGENTS.md
Skill | Description |
|-------|-------------|
| `docx` | Create/edit Word (.docx) |
| `xlsx` | Manipulate Excel (.xlsx) |
| `pptx` | Create PowerPoint presentations |
| `pdf` | Manipulate PDFs |
| `doc-coauthoring` | Document co-authoring |
### Design & Creative
| Skill | Description |
|-------|-------------|
| `algorithmic-art` | Algorithmic
PedroEvangelisAGENTS.mdPedroEvangelis/curr-mvp/AGENTS.md
typecheck # tsc --noEmit
.\compile.ps1 "output/DATA x EMPRESA x VAGA.json" # lint → validate → compile → PDF
.\setup.ps1 # download typst binary to bin/
```
`compile.ps1` automatically runs lint + Zod validation before invoking typst — no need
qitarAGENTS.mdqitar/ooxml-reference/AGENTS.md
type, namespace_uri) tuples;
`CHAPTER_MAP` and `SUBSECTION_MAP` assign ml_type to PDF sections via `section_to_ml`
2. **`_build_schema.py`** — derives ml_type from `PREFIX_MAP` via namespace URIs
qunqin24AGENTS.mdqunqin24/polyglot/AGENTS.md
因为没有这一条它们会掉进 "file" 分支,
然后作为一份上游会拒绝的文档发过去 —— 那就把"还没实现"变成了一次看起来毫不相干的失败。
`TestAudioIsReportedNotSmuggledThroughAsADocument` 钉住了这一点;等音频实现了,是重写那个测试,
而不是删掉它。
**多模态:** 图片和 PDF 在五个协议之间都能转换。内联 base64 是所有协议都能表达的形状,绝不能有
损。远程 URL 只转发给自己会去抓取的协议;Gemini 不会,所以那个配对会被报告出来,除非开了
`FETCH_REMOTE
rangzenAGENTS.mdrangzen/knowledge-project-skills/AGENTS.md
Three tiers of source handling - no hard failures.** (1) Dedicated scripts exist for PDF, DOCX, Excel, CSV, JSON - use them; they extract content and structural metadata. (2) Plain-text formats
redfriesAGENTS.mdredfries/paper2tex/AGENTS.md
path** (e.g., `./figures`): Pass `--figures-dir ` to extraction. The pipeline will reconcile vector (`.pdf`, `.svg`) and high-res raster assets automatically.
### 1. Run preflight check
```bash
python scripts/preflight.py
```
If anything
SanHsienAGENTS.mdSanHsien/book-to-skill/AGENTS.md
SKILL.md`。** 根目錄 `SKILL.md` 是給 Copilot / Amp / Claude Code 安裝的轉換器規格,不是本 fork 的維護索引。維護規則以本檔為準。
- 不提交書籍、PDF、EPUB、產生出來的技能、API key、cookie 或帳號資料。
- 不把他人著作的產生技能公開散布;轉換器本身不附帶任何書籍內容。
- 不把 fork 的 MkDocs 部署到上游網域 `booktoskill.is-a.dev
ScientFactoryAGENTS.mdScientFactory/Scient/AGENTS.md
remains documentation-first. The standalone desktop now
contains permanent Scient-owned project initiation, Sources, PDF, safe text
editing, universal direct file opening, freshness, rich chat math/diagrams/
charts, interactive HTML, browser
siliconstencilAGENTS.mdsiliconstencil/PhantomLogos/AGENTS.md
forward`, `playwright_resize`, `playwright_press_key`, `playwright_drag`, `playwright_upload_file`, `playwright_save_as_pdf`, `playwright_custom_user_agent`, `playwright_click_and_switch_tab`, `playwright_iframe_click`, `playwright_iframe_fill
sshah03AGENTS.mdsshah03/use-citations/AGENTS.md
another question in the same report.
- The first time they give it a PDF, it sets up a small PDF reader in the folder they're
working in. That
sulipapaAGENTS.mdsulipapa/localize-infographic-pdf/AGENTS.md
this when asked to **localize / translate / 本地化 / 汉化 a design-exported PDF**
(infographic, one-pager, flyer, poster — often landscape/multi-panel) while keeping the artwork
identical. It erases the source vector/Type3 text