unified content (audio, video, infographic, slides, etc.) |
| `download_artifact` | Download any artifact (audio, video, pdf, markdown, json) |
| `export_artifact` | Export Data Tables to Google Sheets or Reports to Google Docs
roda em
`app/layout.tsx` e um throw ali é 500 em todas as telas. O PDF de LGPD não leva marca: ele nomeia
o controlador (`organizations.legal_name`) e o DPO.
- **`supabase/baseline.sql
reference uses. It is AGPL-3 and its font exception covers only Postscript and PDF, not SVG.
TeX Gyre Bonum Bold is the same Bookman design under the GUST Font
committed rendered images (`.png`)
- `styles/`: Honkit and print styling
- `exported/`: final deliverables such as `*.pdf` and `*-print.html`
- `tools/book-kit/`: shared Python build/export tooling for both books
Do not treat `_book/`, `_debug
Must not contain consecutive hyphens (`--`)
- Must match the parent directory name
```yaml
# Valid
name: pdf-processing
name: data-analysis
# Invalid
name: PDF-Processing # uppercase not allowed
name: -pdf # cannot start
user
describes what they do — in plain English, or by handing over a PDF, a link, a
script, a transcript — and this skill builds a complete, validated,
security-scanned agent
most 1000 UTF-8 bytes.
- Images are at most 8 MB; audio, video, and PDF files are at most 25 MB.
## Testing
Mock `fetch`; unit tests must not call Meta
verifiable citations.
**Default Format:** LaTeX with BibTeX citations unless otherwise requested.
**Quality Assurance:** Every PDF is automatically reviewed for formatting issues and iteratively improved until visually clean and professional.
**CRITICAL
Calibre 7.6.0 package has an EPUB generation bug (bytes/str mismatch in `container.py`). DOCX and PDF generation work fine. This is a distro packaging issue, not a codebase bug.
- `pypandoc` installs
bridge: version, platform, `capabilities`, `setTheme`, `openExternal`, `savePdf` |
| `src/pdf.js` | renders the calling window to a PDF the user picks a home for; allowlists the options the page may set |
| `src/downloads.js` | gives
with AI and is an experiment. Use at your own responsibility and risk.
- Gesamt-PDF und Einzel-PDFs enthalten weder eine Hinweisseite noch diesen Warntext. Der Hinweis für PDF-Downloads
source file.
- **URLs**: Fetch content, extract title/author/date/text, write to `raw/articles/` or `raw/papers/`; download PDF URLs and run PDF extraction
- **Files**: Read directly, preserve formatting; convert PDFs to markdown when possible