pnpm --filter maister-web db:erd`.
Anything else (PlantUML, draw.io XML, PNG screenshots, PDF) needs an ADR
in `decisions.md` first.
### R2. Mermaid is the only authored diagramming language
Every architectural
106mo agoReviewMountCreate a free account to mount
measures one change.
pnpm kookaburra:run --action memory --project long-device-spike
# Stills (PDF handout or PNG zip, one page per still) through the export path (docs/stills.md).
pnpm kookaburra
stored XSS against the staff
sessions that live there. The case that bites is PDF: browsers open it in a
viewer that runs JavaScript. Three things hold it, each with
102mo agoReviewMountCreate a free account to mount
predates the v0.2 code and roughly half of it was never built (Form 16 PDF parsing, the AIS CSV fallback, the SFT lookup, reconcile H2/M2/L1/L2, the whitelist engine). Every section
MatchFastPath`.
**Python is no longer needed on the compression path** (v0.11.3). ADR
0015 moved PDF text extraction to native Go and ADR 0016 ported
`cordon-filter`, closing both
pdf-toolkit-mcp
## 1. Project Overview
**What**: A write-capable, zero-config, TypeScript-native MCP server for PDF manipulation. Provides 22 tools for reading, modifying, creating, and rendering PDFs — including
SUPPORTED_EXTS` so the walker's globs cannot drift from what the extractor handles. PDF and legacy binary formats are deliberately out of scope.
- **[glean_code/_indexing_walk.py](glean_code/_indexing_walk.py)** — the `--path
debugging cycle:
- **The tenant mounts the kernel's own source** (`#[path] = "../../../kernel/src/image/…"`,
the `pdf-wasm`/`h264diff` pattern), so there is **one** decoder and porting it cannot
regress it. What
glossaries, named-entity lists, register notes. Plain text and Markdown work directly. For .docx, .pdf, .epub, use the **reference-reader** subagent to extract them first.
2. **Build a translation brief
console kept its own list of formats (`ACCEPT`/`BINARY` constants in
`Ingest.jsx`) so every `.pdf` stayed greyed out while the converter was
installed, enabled and loaded; the Models console kept
design web {{COMPANY_NAME}}
## Rôle
Vous produisez les **landing pages**, les **lead magnets** (guides PDF, outils interactifs, quiz, templates) et les **pages statiques** qui vivent hors du CMS du blog
ignorés par Git dans le template ; un fork peut versionner leurs sources, jamais les PDF ni les rendus) |
## La règle qui commande tout : aucune transparence dans le PDF
Ghostscript aplatit
runtime`, `analytics`, `logging`, `loader`, `events`, `storage`, `client`, `security`, `notifications`, `agent-runs`, `otel`, `bridge`, `files-pdf`; `slack` and `teams` are opt-in and outside `full` (`systemprompt/Cargo.toml` is the list)
- `crates/tests/` — separate
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.