@aitne-sh/aitne 0.1.11 → 0.1.13

package/agent-assets/docs/guides/reinstall-cleanly.md

@@ -17,9 +17,7 @@ summary: |
   DB is genuinely corrupt.
 section: reinstall-cleanly
 tags:
-  - guide
   - operations
-  - reinstall
   - migration
   - backup
 status: stable
@@ -30,7 +28,7 @@ ask_examples:
   - Do I need to wipe the DB to upgrade Aitne?
 locale: en-US
 created: 2026-04-25
-updated: 2026-05-28
+updated: 2026-07-01
 keywords:
   - reinstall
   - factory reset
@@ -54,10 +52,12 @@ ui_anchors:
 
 ## You Almost Never Need This
 
-Aitne upgrades through **forward-only schema migrations** applied at
-daemon boot — your DB and context files survive every
+Aitne upgrades itself safely. Each time the background service (the
+**daemon**) starts, it runs **forward-only schema migrations** — small
+scripts that update the database structure without erasing anything. So
+your database and context files survive every
 `npm install -g @aitne-sh/aitne@latest` upgrade. If you're reading this
-because you just upgraded, **don't run the steps below**; restart the
+right after upgrading, **don't run the steps below.** Just restart the
 daemon and let the migration runner do its job.
 
 This guide is the last-resort escape hatch for:
@@ -97,8 +97,8 @@ with different scopes — make sure you're running the one you mean:
    ```bash
    cp -R ~/.personal-agent/context backup/
    ```
-   Context Markdown is what makes the agent know who you are; the DB is
-   recoverable, context is not.
+   Context Markdown is how the agent knows who you are; the DB can be
+   rebuilt, but context cannot.
 3. **Back up the DB** if there's any chance you'll want to recover it
    later:
    ```bash
@@ -115,8 +115,9 @@ with different scopes — make sure you're running the one you mean:
    ```bash
    rm ~/.personal-agent/data/personal_agent.db*
    ```
-   (The `*` glob clears `-shm` / `-wal` left behind by WAL mode. The DB
-   lives inside `data/`, not at the top of the data dir.)
+   (The `*` at the end also clears the `-shm` and `-wal` helper files
+   that SQLite's write-ahead log leaves behind. The DB lives inside
+   `data/`, not at the top of the data dir.)
 3. Start the daemon.
    ```bash
    aitne start
@@ -135,5 +136,5 @@ with different scopes — make sure you're running the one you mean:
 
 - [Backup and Restore](backup-and-restore.md)
 - [Migrate Machines](migrate-machines.md)
-- [Schema Migration](../glossary.md#schema-migration) — the boring
-  alternative to this guide that runs automatically on every boot.
+- [Glossary: Schema Migration](../glossary.md) — the boring alternative
+  to this guide that runs automatically on every boot.

package/agent-assets/docs/guides/setup-wizard.md

@@ -15,8 +15,6 @@ summary: |
   Rules step.
 section: guides
 tags:
-  - core
-  - guides
   - getting-started
   - setup
   - backends
@@ -29,7 +27,7 @@ ask_examples:
   - Where do I paste my API key?
 locale: en-US
 created: 2026-04-25
-updated: 2026-06-10
+updated: 2026-07-01
 keywords:
   - setup
   - wizard
@@ -65,10 +63,10 @@ ui_anchors:
 
 ## Goal
 
-Take a fresh install through the wizard with informed answers at each
-step. The wizard is rerunnable any time at `/setup?mode=update`, which
-jumps straight to the Customize Rules step so you can refine the agent
-without redoing the whole flow.
+Walk a fresh install through the wizard, making an informed choice at
+each step. You can rerun the wizard any time at `/setup?mode=update`,
+which jumps straight to the Customize Rules step so you can refine the
+agent without redoing the whole flow.
 
 ## Prerequisites
 
@@ -77,10 +75,10 @@ without redoing the whole flow.
 
 ## The steps at a glance
 
-The wizard has eight collection steps plus a terminal "Done" screen.
-Three are required (**Basics**, **Vault**, **AI Backend**, plus
-**Customize Rules**); the four integration/messaging steps are
-skippable and re-runnable later.
+The wizard has eight collection steps plus a final "Done" screen. Four
+of them are required — **Basics**, **Vault**, **AI Backend**, and
+**Customize Rules**; the other four (the integration and messaging
+steps) are skippable, and you can run them later.
 
 | # | Step | Required? | What it sets |
 |---|------|-----------|--------------|
@@ -143,7 +141,7 @@ subscription. Anthropic in particular currently prohibits using the
 Claude Agent SDK with a Claude Pro / Max subscription.
 
 Per-process model bindings are seeded with fixed defaults: **Claude
-Sonnet 4.6** (`claude-sonnet-4-6`) for owner-facing work and **Claude
+Sonnet 5** (`claude-sonnet-5`) for owner-facing work and **Claude
 Haiku 4.5** (`claude-haiku-4-5-20251001`) for delegated/simple
 polling. Opus (`claude-opus-4-8`) is registered but not seeded — opt
 in per row. There are no subscription-plan questions. Change the

package/agent-assets/docs/guides/switch-default-backend.md

@@ -14,7 +14,6 @@ summary: |
   different backend.
 section: switch-default-backend
 tags:
-  - guides
   - backends
   - operations
   - routing
@@ -25,7 +24,7 @@ ask_examples:
   - How do I change the main backend the agent uses?
 locale: en-US
 created: 2026-04-25
-updated: 2026-05-28
+updated: 2026-07-01
 keywords:
   - main backend
   - switch backend
@@ -52,9 +51,10 @@ api_endpoints:
 ## Goal
 
 Change which backend the agent uses by default for most work —
-`claude`, `codex`, `gemini`, or `opencode`. This is the broad switch
-that re-seeds every ProcessKey to the new backend's per-tier default
-models. To change just one ProcessKey instead, see
+`claude`, `codex`, `gemini`, or `opencode`. This is the broad switch:
+it re-seeds every ProcessKey (each named job the agent runs, like a DM
+reply or a morning routine) to the new backend's default model for
+that job's tier. To change just one ProcessKey instead, see
 [Change which model handles X](change-which-model-handles-x.md).
 
 ## Prerequisites
@@ -65,10 +65,10 @@ will start failing the moment it routes work there.
 - Register an API key for the backend on `/settings/models`
   (`anthropic` / `openai` / `google`, or one of the cloud-provider
   options for that backend).
-- API keys are the supported path. If no key is set, the daemon falls
-  back to whatever subscription login the backend's CLI already has —
-  the auth-health card on `/settings/models` flags this so you can
-  tell which credential is actually in use.
+- An API key is the supported way in. If no key is set, the daemon
+  falls back to whatever subscription login the backend's CLI already
+  has — the auth-health card on `/settings/models` flags this so you
+  can tell which credential is actually in use.
 
 ## Steps
 

package/agent-assets/docs/guides/use-an-existing-obsidian-vault.md

@@ -14,9 +14,7 @@ summary: |
   through Adopt vs Migrate vs Split.
 section: use-an-existing-obsidian-vault
 tags:
-  - guides
   - wiki
-  - obsidian
   - knowledge
 status: stable
 ask_examples:
@@ -26,7 +24,7 @@ ask_examples:
   - Why is Split greyed out?
 locale: en-US
 created: 2026-05-12
-updated: 2026-06-07
+updated: 2026-07-01
 keywords:
   - external vault
   - existing obsidian vault
@@ -58,11 +56,11 @@ process_keys:
 
 ## Goal
 
-Aitne can point the wiki subsystem at a vault you already maintain
-in Obsidian, instead of creating its own at `$PA_DATA_DIR/wiki`.
-This is "external mode", and the setup wizard walks you through
-three decisions: **probe**, **classify**, then **Adopt vs Migrate
-vs Split**.
+Aitne can point the wiki at a vault you already keep in Obsidian,
+instead of creating its own at `$PA_DATA_DIR/wiki`. This is called
+"external mode". The setup wizard walks you through three steps:
+it **probes** the folder you pick, **classifies** what it finds,
+then asks you to choose **Adopt vs Migrate vs Split**.
 
 ## Step 1 — Open Settings → Wiki
 
@@ -110,9 +108,9 @@ levels:
   an existing wiki layout is detected, the wizard will surface the
   Adopt / Migrate decision in Step 3.
 
-The **Use this folder** button is greyed out while the banner
-is an error so you cannot fire a workspace-create that the daemon
-already knows will fail.
+The **Use this folder** button stays greyed out while the banner
+shows an error, so you cannot start a workspace-create that the
+daemon already knows will fail.
 
 ## Step 2 — Probe Classifies the Vault
 

package/agent-assets/docs/pages/activity.md

@@ -0,0 +1,76 @@
+---
+schema_version: 1
+slug: pages/activity
+title: "Agent Log Page"
+id: page-activity
+aliases:
+  - agent log page
+  - activity page
+  - log page
+  - conversations page
+category: pages
+summary: |
+  The Agent Log is the historical record of everything the agent has seen
+  or done — invocations (Events), daemon output (System Logs), completed
+  chat threads (Conversations), and proactive outbound messages
+  (Notifications).
+tags:
+  - audit
+  - notifications
+  - operations
+status: stable
+ask_examples:
+  - What can I do on the Agent Log page?
+  - Where do I see why a routine failed?
+  - Where do I read a full conversation transcript?
+  - Where are the daemon's system logs?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - agent log
+  - activity
+  - events
+  - system logs
+  - conversations
+  - notifications
+related:
+  - features/operations/activity-and-conversations
+  - features/operations/notifications
+  - features/operations/backend-routing
+  - pages/analytics
+ui_anchors:
+  - /activity
+  - /conversations
+---
+
+# Agent Log Page
+
+The `/activity` page is the agent's audit trail — a permanent record you can
+look back on of everything the agent has seen or done, split across four tabs.
+
+## What you can do here
+
+- **Events** — one row each time the agent ran (an "invocation"): what
+  triggered the run, which backend served it, and whether it succeeded or
+  failed. This is where you diagnose a routine that didn't do what you expected.
+- **System Logs** — output from the daemon itself: startup messages, errors,
+  and warnings from the Hono server.
+- **Conversations** — completed chat threads from every messaging platform,
+  each of which you can open as a full transcript.
+- **Notifications** — messages the agent sent you on its own, without you
+  asking.
+
+## Where to go deeper
+
+- [Activity & Conversations](../features/operations/activity-and-conversations.md)
+  — the audit timeline and per-session transcripts in detail.
+- [Notifications](../features/operations/notifications.md) — when and how
+  the agent reaches out.
+- [Backend routing](../features/operations/backend-routing.md) — how to
+  read the "which backend served it" column.
+
+## Related
+
+- [Analytics page](analytics.md) — the aggregated view of this same data.
+- [Overview page](overview.md) — the recent-events feed is a slice of Events.

package/agent-assets/docs/pages/agents.md

@@ -0,0 +1,78 @@
+---
+schema_version: 1
+slug: pages/agents
+title: "Agents Page"
+id: page-agents
+aliases:
+  - agents page
+  - agents hub
+  - routines page
+category: pages
+summary: |
+  The Agents page is the hub for everything that runs on your behalf — the
+  shipped System routines and your own User Agents. Each has a schedule, a
+  rulebook, limits, and recent health you can review and edit here.
+tags:
+  - agents
+  - routines
+  - autonomous
+status: stable
+ask_examples:
+  - What can I do on the Agents page?
+  - How do I create a new agent?
+  - Where do I edit a routine's rulebook or schedule?
+  - How do I see how often an agent fails?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - agents
+  - routines
+  - schedule
+  - rulebook
+  - system agents
+related:
+  - pages/tasks
+  - concepts/routines
+  - features/routines/morning-routine
+  - features/routines/custom-routines
+ui_anchors:
+  - /agents
+---
+
+# Agents Page
+
+The `/agents` page lists each automation as its own named thing you can
+open and edit — not as a one-off item waiting in a line. **System** agents
+are the routines that ship with the app (synthesis, monitoring,
+maintenance); **User** agents are the ones you create yourself.
+
+## What you can do here
+
+- **Browse, search, and filter** every agent, grouped by System vs. User.
+- **Check health at a glance** — last run time and 7-day error rate sit on
+  each card.
+- **Create a new Agent** with the **New Agent** button (name, schedule,
+  rulebook, and limits).
+- **Open an agent** to edit its rulebook and journal rules, adjust its
+  schedule window and cost/turn limits, and read recent executions.
+- **Stop a System agent** (they can't be deleted) or fully manage your own
+  User agents.
+
+What is about to run next lives on the [Tasks](../pages/tasks.md) queue,
+not here.
+
+## Where to go deeper
+
+- [Routines](../concepts/routines.md) — what runs, when, and why.
+- [Morning Routine](../features/routines/morning-routine.md) ·
+  [Evening Review](../features/routines/evening-review.md) ·
+  [Weekly Review](../features/routines/weekly-review.md) ·
+  [Activity Scan](../features/routines/activity-scan.md) — the built-ins.
+- [Custom routines](../features/routines/custom-routines.md) — authoring
+  your own User agents.
+
+## Related
+
+- [Tasks page](tasks.md) — the run queue and scheduled DMs.
+- [Settings page](settings.md) — global hours and notification policy.

package/agent-assets/docs/pages/analytics.md

@@ -0,0 +1,70 @@
+---
+schema_version: 1
+slug: pages/analytics
+title: "Analytics Page"
+id: page-analytics
+aliases:
+  - analytics page
+  - cost page
+  - metrics page
+category: pages
+summary: |
+  The Analytics page shows usage and spend trends — Cost aggregates
+  per-run USD by backend and process across days, weeks, and months;
+  Metrics shows operational health like activity volume and error rates.
+  All from the local database, nothing sent externally.
+tags:
+  - cost
+status: stable
+ask_examples:
+  - What can I do on the Analytics page?
+  - Where do I see how much the agent is spending?
+  - How do I check the agent's error rate?
+  - Is my usage data sent anywhere?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - analytics
+  - cost
+  - spend
+  - metrics
+  - error rate
+related:
+  - concepts/costs-and-quotas
+  - features/operations/cost-tracking
+  - features/operations/backend-routing
+  - pages/activity
+ui_anchors:
+  - /analytics
+---
+
+# Analytics Page
+
+The `/analytics` page takes the raw event log and turns it into usage and
+spend trends, laid out across two tabs.
+
+## What you can do here
+
+- **Cost tab** — per-run USD spend (token counts × backend pricing),
+  totaled by day, week, and month and split out by backend and process.
+  This is where you answer "what is this costing me, and where."
+- **Metrics tab** — how the agent is running: activity volume, a breakdown
+  of what ran, error rates, and how many notifications went out.
+
+All figures come from the daemon's local SQLite database — nothing is sent
+to any external service.
+
+## Where to go deeper
+
+- [Costs & quotas](../concepts/costs-and-quotas.md) — how spend and
+  budgets work.
+- [Cost tracking](../features/operations/cost-tracking.md) — how per-run
+  cost is computed and attributed.
+- [Backend routing](../features/operations/backend-routing.md) — why one
+  backend costs more than another.
+
+## Related
+
+- [Agent Log page](activity.md) — the per-event detail behind these totals.
+- [Settings → Models](settings.md) — change which model handles which tier.

package/agent-assets/docs/pages/browser.md

@@ -0,0 +1,73 @@
+---
+schema_version: 1
+slug: pages/browser
+title: "Browser Page"
+id: page-browser
+aliases:
+  - browser page
+  - browser hub
+category: pages
+summary: |
+  The Browser page is the home for everything the agent does with a
+  browser: reading your existing Chrome history, driving a dedicated
+  sandboxed Chromium, the experimental purchase-confirmation flow, and the
+  browser tasks it has run.
+tags:
+  - autonomous
+status: stable
+ask_examples:
+  - What can I do on the Browser page?
+  - How do I let the agent read my browsing history?
+  - Where do I see browser tasks the agent ran?
+  - What is the managed Chromium for?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - browser
+  - chromium
+  - browsing history
+  - automation
+  - browser tasks
+related:
+  - features/operations/managed-chromium
+  - features/operations/browser-tasks
+  - features/integrations/browser-history
+ui_anchors:
+  - /browser
+  - /browser-tasks
+---
+
+# Browser Page
+
+The `/browser` page brings every browser feature together in one place.
+Each one has its own on/off switch and settings.
+
+## What you can do here
+
+- **Browser History** — let the agent read your existing Chrome history so
+  it can build research clusters and add revisit nudges to the morning
+  digest.
+- **Browser Automation** — set up the dedicated, sandboxed Chromium (a
+  separate, isolated browser) that the agent drives for sign-in (OAuth)
+  sites and task slots. A hostname denylist limits which sites it may open.
+- **Purchase Confirmations** — the experimental B-4 flow. Before the agent
+  can go through with a purchase, it must use a single-use token that is
+  sent to you by DM. This is off by default.
+- **Browser tasks** — review the tasks the agent has run, including any
+  waiting for your confirmation, one click away at `/browser-tasks`.
+
+## Where to go deeper
+
+- [Managed Chromium](../features/operations/managed-chromium.md) — the
+  sandboxed browser the agent drives.
+- [Browser tasks](../features/operations/browser-tasks.md) — how a browser
+  task is requested, run, and confirmed.
+- [Browser history](../features/integrations/browser-history.md) — reading
+  from your existing Chrome.
+
+## Related
+
+- [Connections page](connections.md) — where other integrations are linked.
+- [Settings page](settings.md) — Browser History and Automation also have
+  settings panels.

package/agent-assets/docs/pages/chat.md

@@ -0,0 +1,67 @@
+---
+schema_version: 1
+slug: pages/chat
+title: "Chat Page"
+id: page-chat
+aliases:
+  - chat page
+  - dashboard chat page
+category: pages
+summary: |
+  The Chat page is an in-dashboard conversation with your agent — the same
+  agent you reach from Telegram, Slack, Discord, and WhatsApp, just from a
+  browser tab. Send messages, run bang commands, and revisit past sessions.
+tags:
+  - messaging
+status: stable
+ask_examples:
+  - What can I do on the Chat page?
+  - Can I talk to the agent from the dashboard?
+  - Do bang commands work in the dashboard chat?
+  - Where do I find an old chat session?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - chat
+  - conversation
+  - messaging
+  - dashboard chat
+related:
+  - features/messaging/dashboard-chat
+  - features/messaging/overview
+  - features/messaging/bang-commands
+ui_anchors:
+  - /chat
+---
+
+# Chat Page
+
+The `/chat` page lets you talk to your agent right inside the dashboard.
+It uses the same agent, memory, and tools as the messaging apps — so
+anything you could send the agent in a DM, you can type here.
+
+## What you can do here
+
+- **Message the agent.** Ask questions, give instructions, or follow up on
+  something it did. Replies stream back word by word, in real time.
+- **Run bang commands.** `!help`, `!ingest`, `!compile`, `!ask`, and the
+  rest work exactly as they do on messaging apps.
+- **Revisit past sessions.** Switch to an earlier conversation to re-read
+  what was said; completed threads are also visible under
+  [Agent Log → Conversations](activity.md).
+
+## Where to go deeper
+
+- [Dashboard Chat](../features/messaging/dashboard-chat.md) — the Chat
+  page in detail.
+- [Messaging Overview](../features/messaging/overview.md) — the same agent
+  from Telegram, Slack, Discord, and WhatsApp, plus how pairing and voice
+  notes work.
+- [Bang commands](../features/messaging/bang-commands.md) — the full `!`
+  command reference.
+
+## Related
+
+- [Agent Log page](activity.md) — where completed conversations are archived.
+- [Connections page](connections.md) — link the messaging apps.

package/agent-assets/docs/pages/connections.md

@@ -0,0 +1,78 @@
+---
+schema_version: 1
+slug: pages/connections
+title: "Connections Page"
+id: page-connections
+aliases:
+  - connections page
+  - integrations page
+  - connections hub
+category: pages
+summary: |
+  The Connections page is the hub for linking outside services to the
+  agent — messaging apps, git repositories, note sources (Obsidian and
+  Notion), calendar, mail, and MCP servers. Each has its own sub-page.
+tags:
+  - integrations
+status: stable
+ask_examples:
+  - What can I do on the Connections page?
+  - How do I connect my calendar or mail?
+  - Where do I link Obsidian or Notion?
+  - How do I add an MCP server?
+locale: en-US
+created: 2026-07-01
+updated: 2026-07-01
+keywords:
+  - connections
+  - integrations
+  - messaging
+  - calendar
+  - mail
+  - mcp
+related:
+  - features/messaging/overview
+  - features/integrations/calendar
+  - features/integrations/mail
+  - concepts/skills
+ui_anchors:
+  - /connections
+---
+
+# Connections Page
+
+The Connections area is where you connect outside services — your chat
+apps, calendar, notes, and more — to the agent. It opens on the
+**Messaging** tab, and a sidebar on the left switches between the
+connection types.
+
+## What you can do here
+
+- **Messaging** (`/connections/messaging`) — pair Telegram, Slack, Discord,
+  and WhatsApp so you can DM the agent.
+- **Repositories** (`/connections/repositories`) — register git repos to
+  watch (then tune them on the [Git page](../pages/git.md)).
+- **Notes** (`/connections/notes`) — connect your personal Obsidian vault
+  and Notion.
+- **Calendar** (`/connections/calendar`) — link Google Calendar.
+- **Mail** (`/connections/mail`) — connect a mail account.
+- **MCP Servers** (`/connections/mcp`) — attach Model Context Protocol
+  servers that give the agent more tools.
+
+## Where to go deeper
+
+- [Messaging Overview](../features/messaging/overview.md) — supported apps
+  and pairing.
+- [Calendar](../features/integrations/calendar.md) ·
+  [Mail](../features/integrations/mail.md) ·
+  [Git](../features/integrations/git.md) ·
+  [Obsidian](../features/integrations/obsidian.md) ·
+  [Notion](../features/integrations/notion.md) — per-integration detail.
+- [Skills](../concepts/skills.md) — how MCP servers expand what the agent
+  can do.
+
+## Related
+
+- [Git page](git.md) — configure repositories after registering them here.
+- [Settings page](settings.md) — the agent's own primary vault is managed
+  from Settings → Management.