Files changed:
- CHANGES.md
- VERSION
- docs/how-to/ingest-a-source.md
- docs/reference/commands/raw-material-and-uploads.md
- raw/CONTRACT.md
- tools/CONTRACT.md
- tools/chemenu/commands/raw/derive.py
- tools/chemenu/filters/pdf.py
- tools/chemenu/filters/pdf_layout.py
- tools/chemenu/tests/binary_formats.py
- tools/chemenu/tests/test_raw_derive_binary.py

Co-Authored-By: Claude Code <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SnAJ7Z3CpVD3PRbN73QtU2

2026-10-14 20:23:07 +02:00

back-up-to-a-synced-folder.md

export: export run writes selected kb/ pages as a derived, read-only version into configured target folders (markdown and command handlers, manifest, after_publish), raw accept also refuses manifest-listed files, doctor checks the section (#214 ); 10.0.0-beta.13

2026-10-13 05:53:48 +02:00

browse-with-obsidian.md

feat: docs/ carries the operator's how-to guides, a configuration reference and the architecture and telemetry explanations carved out of README.md, INSTALL.md, INSTALL-MCP.md and EVALS.md; the last two retired (#190 )

2026-10-07 05:13:56 +02:00

capture-repository-docs.md

2026-10-07 05:13:56 +02:00

collect-from-a-synced-folder.md

docs: collect how-to records what OneDrive on Windows does to a phone upload (online-only placeholder, original times kept, so settle_seconds does not hold it back), export how-to says a tenant found and used the .md files (#213 E3, #214 F1)

2026-10-13 13:51:31 +02:00

export-guidelines.md

2026-10-13 05:53:48 +02:00

export-to-a-synced-folder.md

2026-10-13 13:51:31 +02:00

fetch-pages-behind-a-login.md

raw: fetch behind a login - raw login stores the user's session as a secret outside the tree, fetch.routes sends a host through it or to an external fetcher, a handler's render goes through the optional browser component (preflight --with browser) (#169 ); 10.0.0-beta.8

2026-10-11 20:50:23 +02:00

ingest-a-source.md

2026-10-14 20:23:07 +02:00

maintain-the-wiki.md

docs: maintain-the-wiki names the synced-folder backup for an instance without a git server, and what a failed machine-data backup after a push means (#186 close)

2026-10-12 19:58:14 +02:00

query-the-wiki.md

2026-10-07 05:13:56 +02:00

README.md

2026-10-13 05:53:48 +02:00

run-the-weekly-review.md

settings: the four per-feature files .wikitool-remotes/-telemetry/-upload/-tasks.json move into the three tiers and disappear; new command config migrate; 10.0.0-beta.1 (#208 )

2026-10-11 06:54:35 +02:00

set-up-the-mcp-server.md

intake: several configured intake channels (mcp, folder, command) all end in quarantine/ (was mcp-upload/), and upload collect pulls the synced-folder and command channels into it (#213 ); 10.0.0-beta.11

2026-10-12 22:04:18 +02:00

trace-and-score-sessions.md

settings: the four per-feature files .wikitool-remotes/-telemetry/-upload/-tasks.json move into the three tiers and disappear; new command config migrate; 10.0.0-beta.1 (#208 )

2026-10-11 06:54:35 +02:00

troubleshoot.md

settings: the four per-feature files .wikitool-remotes/-telemetry/-upload/-tasks.json move into the three tiers and disappear; new command config migrate; 10.0.0-beta.1 (#208 )

2026-10-11 06:54:35 +02:00

update-an-instance.md

docs: the update how-to says the 10.0.0 settings migration is owed per checkout, so a second clone runs config migrate after its pull (#208 close)

2026-10-11 10:23:05 +02:00

use-1password-for-settings.md

docs: settings pull-through after #207 - secrets file path on Windows and with XDG in the 1Password how-to; 1Password approval is per terminal session, not per value

2026-10-10 18:10:07 +02:00

write-a-site-handler.md

sites: raw fetch --scope fetches a page and the subpages its site handler offers as one bundle in incoming/<stem>/, with robots.txt, max_pages and pause_seconds; 10.0.0-beta.6

2026-10-11 18:36:10 +02:00

README.md

How-to guides

Task-oriented pages for the operator of an instance: one goal per page - upgrade an instance, set up the MCP read server, run a weekly review - with the steps a human takes and the sentence handed to the agent. A how-to guide never retells an agent's procedure: the steps an agent runs live in instructions/, and a guide here says what to prepare, what to ask for and where the agent stops.

Working with the wiki:

Ingest a source - a file, a folder, a web page, or a document submitted through the MCP server
Write a site handler - teach raw fetch how a website serves its pages
Fetch pages behind a login - with your own session, through a browser, or through a fetch program of your own
Capture documentation from a git repository - and keep it current
Export guidelines to captured repositories
Export to a synced folder - a read-only version of selected pages for a reader outside the repository, OneDrive and Microsoft 365 included
Query the wiki - asking, and searching yourself
Maintain the wiki - lint, provenance, and the gates on the way to a publish
Run the weekly review - commitments in a task tracker against the project pages
Browse the wiki in Obsidian - and how pages are named

Running the instance:

Update an instance - to a newer release of the stack
Troubleshoot an instance
Set up the MCP read server - for a consumer that is not a terminal on this machine
Trace and score agent sessions
Use 1Password for settings - secrets and private values out of the tree, on the desktop and in CI
Collect documents from a synced folder - a folder OneDrive and the like fill, or a program of your own, as an intake channel into the quarantine
Back up to a synced folder - OneDrive and the like as the backup of an instance without a git server, and restoring from it

Installing a new instance is INSTALL.md at the repository root.