brain-of-reese/README.md

# 🧠 Brain of Reese

A chippy, honest **RAG chatbot** over your `~/Homelab` and `~/Deployments`
projects. Point it at your notes — markdown, YAML, JSON, Python, plain
text — ask it anything, and it retrieves the relevant chunks with
**hybrid search** (pgvector cosine ∪ Postgres full-text search, fused with
Reciprocal Rank Fusion), feeds the **whole relevant document** to a
**self-hosted LLM** (`turbo` via `https://aipi.reeseapps.com/v1`), and
streams a grounded answer back.

If it doesn't have notes for your question, it admits it: *"I haven't done
anything like that"* — plus suggestions for what it **does** know.

> **Updated your notes?** Re-run the import — it's idempotent and only
> re-embeds what changed:
> ```bash
> uv run python -m scripts.import_docs --prune
> ```
> See [Updating the documents](#updating-the-documents) for details.

---

## Quick Start

```bash
# 1. Dependencies
uv sync

# 2. Configure (copy and edit)
cp .env.example .env
# Set BOR_LLM_API_KEY — your aipi key (falls back to $AIPI_KEY)

# 3. Start the database
podman compose up -d db
podman compose ps          # wait until "healthy"

# 4. Apply migrations & import
uv run alembic upgrade head
uv run python -m scripts.import_docs

# 5. Run the app
uv run uvicorn app.main:app --reload
# → http://localhost:8000  (chat)
```

**Stack:** FastAPI · Pydantic v2 · SQLAlchemy 2 · Alembic · pgvector ·
vanilla HTML/CSS/JS (no CDN) · Playwright E2E

---

## Pages at a glance

The app is a single-page application — all views load from `/` with a
client-side router. Direct URLs deep-link to the matching view.

| Page | URL | Who can see it |
|------|-----|----------------|
| **Chat** | `/` | Everyone (token gate for non-shared) |
| **RAG / Knowledge base** | `/sources.html` | Admin only |
| **Git sources** | `/git-sources.html` | Admin only |
| **Tuning** | `/tuning.html` | Admin only |
| **Saved chats** | `/history.html` | Admin only |
| **Access tokens** | `/tokens.html` | Admin only |
| **Document viewer** | `/document.html?source=X&path=Y` | Token users & admins |
| **Login** | `/login.html` | Everyone |
| **Shared chat** | `/shared/<token>` | Everyone (anonymous) |
| **Edit doc** | `/doc-edit.html` | Admin only |

### Chat (`/`)

Ask questions; answers stream in with **source chips** that cite the exact
documents used. Clicking a chip opens the document in an **almost-fullscreen
modal** on the same page (no new tab). The onboarding suggestion chips
follow the opening questions of the 3 most recent saved chats (each
chat's first question — a follow-up never chips, phase 103) — on a
fresh deployment they seed from `BOR_SUGGESTIONS`.

### RAG / Knowledge base (`/sources.html`)

The indexed document list. The *Path* column opens each document in the same
**almost-fullscreen modal**. Admin-only — anonymous visitors see a sign-in
gate instead.

### Git sources (`/git-sources.html`)

The admin-managed source registry: add or remove git repositories, upload
source archives (`.tar`, `.tar.gz`, `.tgz`, `.zip`), and register local
directories — all in one table with a `kind` discriminator. No `.env`
editing, no restart. The **Sync sources** button clones/pulls/walks every
source and re-imports in one click.

### Tuning (`/tuning.html`)

Set global instructions that are injected into the system prompt of every
chat turn. Add notes like *"be more concise"* or *"assume I'm on NixOS"*.
List, edit, or remove them here — no conversation required.

### Saved chats (`/history.html`)

Every conversation is saved automatically. Click a title to return to that
chat. Admin-only view.

### Access tokens (`/tokens.html`)

Generate tokens and hand them out — a token opens chat, the answers, and
the documents they cite. Shared chats stay open to everyone.

### Shared chat (`/shared/<token>`)

A read-only snapshot of any conversation, shareable via a link. Open to
everyone — no login or token required.

### Edit doc (`/doc-edit.html`)

Review and adjust an AI-generated answer before committing it to the docs
repository. Admin-only flow page.

---

## Admin & Sign-in

Brain of Reese has exactly **one account: the admin (you)**. Signing in
unlocks the **full Sources catalog**, the **answer-tuning** controls, and
**token management**. **Shared chats** are the only content that stays open
to anonymous visitors.

### Setup (one-time)

```bash
python -c 'import secrets;print(secrets.token_hex(32))'   # → paste into .env
```

```env
BOR_ADMIN_PASSWORD=your-password        # plaintext — homelab scope, by design
BOR_SESSION_SECRET=<the hex from above> # signs the session cookie
```

**Fail-loud:** while either variable is empty the app refuses to start,
naming the missing one(s).

### How authentication works

| Endpoint | Purpose |
|----------|---------|
| `POST /api/login` | Password → signed `bor_session` cookie |
| `POST /api/logout` | Clears session cookie |
| `GET /api/whoami` | `{"authenticated": bool, "role": "admin"|"user"|"anonymous"}` |
| `POST /api/token-auth` | Token (`bor_…`) → same cookie, role `user` |

The cookie uses `same_site="lax"`, `https_only` off — **no HTTPS
enforcement** (homelab HTTP; the cookie is single-admin convenience, not a
cloud boundary). Max age is `BOR_SESSION_MAX_AGE` (default `43200` = 12 h).

### Access tokens

The admin can hand out access without sharing the admin password.

1. **Generate:** sign in → **Tokens** in the navbar. Type a label and hit
   **Generate** — a `bor_` + 32-hex token appears in the *shown once* block.
   Copy it now: only its SHA-256 hash is stored.
2. **Use:** the holder enters the token at the in-app gate. The browser
   caches it in `localStorage["bor.token"]` for silent re-auth on reloads.
3. **Revoke:** click **Revoke** on the token's row. Revocation is immediate.

A token user can chat and open documents — nothing else. The Sources
catalog, git sources, tuning, and history stay admin-only.

---

## Using the App

### Updating the documents

The knowledge base is refreshed by **re-running the import**. It is
idempotent and delta-based (sha256 per file), so a refresh after a normal
editing session takes seconds:

```bash
# After editing/adding/removing notes:
uv run python -m scripts.import_docs                 # re-index what changed
uv run python -m scripts.import_docs --prune         # also drop deleted files

# Point it at extra directories (repeatable):
uv run python -m scripts.import_docs --source ~/SomeOtherDocs
```

With **git-based sources** each run first pulls the latest commits of your
repos, so this same command is the whole update loop: *commit → re-run*.

Then check the **Sources** page (`http://localhost:8000/sources.html`):
the *documents* / *chunks* counters and *last indexed* timestamp should
reflect the new files.

- The import prints one line per file and ends with a greppable summary,
  so it is safe to run from cron or after every commit.
- Indexed formats: **md, markdown, txt, yaml, yml, json, py**, the Podman
  quadlet family (**container, network, volume, image, pod, kube, swap,
  os, endpoint**), and **j2** Jinja templates (case-insensitive).
- Hidden files and common cache dirs (`.venv`, `node_modules`, `.git`,
  `__pycache__`, etc.) are skipped.
- Non-markdown files get format-aware chunking (YAML keys, JSON keys,
  Python classes/defs).

### Git-based sources

Rather than pointing the import at local folders, point it at
**git repositories**. The admin **Git sources page** (`/git-sources.html`)
is the primary management surface — add or remove repositories there and
the list is stored in Postgres.

```env
# .env — the fallback list (fresh setups, or until the admin page
# stores a source; the page becomes the source of truth)
BOR_GIT_SOURCES=https://git.reeseapps.com/reese/homelab.git,git@github.com:reese/deployments.git
BOR_SOURCES_DIR=~/bor-sources   # default; each repo lands in <dir>/<repo-name>/
```

- Auth is whatever the machine supplies — HTTPS via the OS credential
  helper, or SSH via your key; no credentials are stored in the app.
- Every run **clones** (first time, shallow `--depth 1`) or **pulls**
  (`git pull --ff-only`) each repo, then indexes the checkout.
- A failed sync aborts the run — no partial junk.

### Archive upload sources

Upload a `.tar`, `.tar.gz`, `.tgz`, or `.zip` archive to make it a source.
The form on the Git sources page — and the `POST /api/git-sources/upload`
route behind it — accepts archives and unpacks + scans them immediately.

- Re-uploading the same filename replaces the source **in place** (no
  second folder, no duplicate row).
- Zip-bomb guard: absolute member paths, `..` traversal, and symlinks
  escaping the unpack folder are rejected.
- Archives unpack under `BOR_UPLOAD_DIR` (default `~/bor-sources/uploads`).
- `BOR_UPLOAD_MAX_MB` (default 512) caps both compressed and extracted size.

### Local directory sources

A plain directory can be a first-class source too. It shares the git
sources' **one table** (the `git_sources` registry with a `kind`
discriminator: `git` | `local`), **one admin page**, and **one Sync button**.

- Register via API: `POST /api/git-sources` with `{"kind": "local", "path": …}`.
- Sync walks it directly — no clone, no checkout copy.
- The directory is re-verified to exist at sync time; a missing directory
  fails the run loudly.

### Sync from the UI

The **Sync sources** button on the **Sources** page (admin only) runs the
whole git-source refresh in one click:

1. **clone/pull + walk** every configured source — git repos and local
   directories, including uploaded archives.
2. **re-import with prune** — files deleted upstream leave the index.
3. **regenerate the KB overview** (the `<knowledge_base>` outline every
   chat turn injects) — only when the import changed the knowledge base.

- States: the button shows **Syncing…** with a spinner while polling
  `GET /api/sync/status` every 2 s. No client-side timeout — a clone +
  embed can legitimately take minutes.
- One sync at a time: a second trigger while a run is in flight gets a
  `409`.

---

## Thinking

The self-hosted `turbo` model reasons before it answers. That reasoning is
streamed with the turn as `thinking` SSE events and shown in a
**collapsible "Thinking" block** above the answer bubble: it opens and
fills in live while the model thinks, tucks itself away the moment the
first answer token lands, and stays click-toggleable afterwards.

To hide it, set `BOR_STREAM_THINKING=0`.

---

## Agent document tools (ls + read + grep)

Retrieval only puts the top documents in context. When an answer depends on
a file a note *references* ("the exact JSON shape is in
`example-record-file.json`"), the model can extend its own context with
three server-side tools — on **grounded** (high-relevance) turns only:

| Tool | Description |
|------|-------------|
| **`ls`** | Lists every indexed document (`source: X \| path: Y \| title: Z`); pass a source name as `path` to list one source |
| **`read(path)`** | Appends the **full** text of one indexed document to context (never truncated); `path` is the combined `source/path` string |
| **`grep(pattern, path?)`** | Searches indexed documents for an exact string (case-insensitive fixed substring); returns up to 20 `source/path:line: text` matches; an optional `path` limits to one document |

Each call is executed against Postgres only (no extra LLM round trip) and
streamed as an SSE `tool` frame. In the chat, each call shows a transient
**calling-tool status** alongside "thinking" (the send button keeps its
busy state — "Stop" — for the whole turn).

The model may call tools as many times as needed, bounded by a round cap:

| Env | Default | Meaning |
|-----|---------|---------|
| `BOR_AGENT_MAX_ROUNDS` | `10` | hard cap on agent tool rounds per grounded turn (0 = no tools, the kill switch) |

---

## Tuning your answers

*Admin-only* — sign in first. If an answer isn't quite right — too chatty,
wrong assumption, missing context — **tune** Brain right there:

1. Press **"Tune"** in the meta row under any completed answer (deflected
   ones included).
2. Type a short instruction (1–2000 chars), e.g. *"be more concise"* or
   *"assume I'm on NixOS"*, and **Save**.

The note is stored in Postgres (`steering_notes`) and read into the
**system prompt of every subsequent chat turn** as a `<tuning>` section
(numbered, oldest first, capped at `BOR_STEERING_MAX_CHARS` chars — default
8000, overflow marked `[…truncated…]`).

List or remove notes from the **"Tuning"** button in the chat header (count
badge, newest-first, per-note delete). The API is stateless JSON if you
prefer curl:

```bash
curl -s localhost:8000/api/steering                                  # list
curl -s -X POST localhost:8000/api/steering \
     -H 'Content-Type: application/json' -d '{"note": "be more concise"}'
curl -s -X DELETE localhost:8000/api/steering/<note-id>              # remove
```

---

## How retrieval works (hybrid)

Every question is embedded and also lexically tokenized (OR-joined, English
stemming) and searched **twice** against Postgres:

1. **Vector** — pgvector cosine top-N (default `BOR_HYBRID_VECTOR_CANDIDATES=100`)
2. **Lexical** — a stored `tsvector` (GIN-indexed) matched with `to_tsquery`,
   top-N by `ts_rank` (default `BOR_HYBRID_LEXICAL_CANDIDATES=30`)

The two ranked lists are fused with **Reciprocal Rank Fusion**
(`score = Σ 1/(k + rank)`, `BOR_RRF_K=60`) — a chunk in both lists scores
nearly double, which lets a name-your-tool question ("gitlab") find its own
document even when the question embeds close to generic templates.

The **honesty gate** (A8) then answers (HIGH) when the best cosine is ≥
`BOR_RELEVANCE_THRESHOLD` (default `0.62`) **or** at least one chunk matched
lexically (`fts_hits > 0`) — it deflects (LOW) only when *both* signals are
absent. The top `BOR_TOP_N_DOCS` full documents are what the LLM sees.

---

## Document summaries (non-markdown)

Raw yaml/json/py/txt embeds badly — flags and keys are not language, so
retrieval can miss exactly the documents that are all configuration. At
import time, every **non-markdown** A9 document is summarized by the aipi
`lite` model (`BOR_LLM_SUMMARY_MODEL`, default `lite`):

* The summary is stored on `documents.summary` **and** indexed as one extra
  embedded chunk (`chunks.is_summary`, position −1), so hybrid search has a
  natural-language target to hit instead of the raw text.
* The last line is a **code-deterministic** pointer — `Source: <source>/<path>`
  — appended by the app, never model-generated.
* The model only sees the first `BOR_SUMMARY_MAX_CHARS` (default 12000)
  characters; overflow is cut and marked with `[…truncated…]`.

Summary generation is best-effort: if `lite` fails, the document is still
indexed (without a summary), the failure is logged, and counted in the
import summary line (`summaries=N summary_errors=N`).

---

## Caching & deploys

A deploy is a commit — and the browser must see it **without a hard refresh**.
One Starlette middleware (`app/core/caching.py`) applies the rule at the
transport layer:

- **HTML pages** ship `Cache-Control: no-cache`, **no `etag`**, **no
  `last-modified`** — each visit always gets a fresh 200 body.
- **Assets are versioned and cached for a year.** Pages reference their
  CSS/JS with a token (`/assets/styles.css?v=<token>`), and every
  `/assets/*` response ships `Cache-Control: public, max-age=31536000, immutable`.
- **The token is the deploy.** In a git checkout it is the short SHA of
  `HEAD` — so every commit/deploy flips the token and the versioned asset
  URLs change with it. A checkout without `.git` falls back to a stable
  content hash of the `frontend/` tree.

No CDN, no new services, no build-step change: the middleware rewrites the
asset references of the known pages in flight.

> **Deploy note:** the very first deploy onto this scheme needs one normal
> page visit, so the browser revalidates the HTML once and starts requesting
> the versioned assets; every commit after that is picked up automatically.

---

## Configuration reference

| Env | Default | Meaning |
|-----|---------|---------|
| `BOR_APP_NAME` | `Brain of Reese` | Display name everywhere (page titles, header brand, status labels) |
| `BOR_INPUT_PLACEHOLDER` | `Ask me anything…` | Chat composer placeholder |
| `BOR_FOOTER_TEXT` | `Powered by self-hosted models` | Footer line on every page |
| `BOR_DATABASE_URL` | local compose URL | SQLAlchemy URL (psycopg) |
| `BOR_LLM_BASE_URL` | `https://aipi.reeseapps.com/v1` | OpenAI-compatible endpoint |
| `BOR_LLM_API_KEY` | — (falls back to `$AIPI_KEY`) | aipi API key |
| `BOR_LLM_CHAT_MODEL` | `turbo` | Chat model |
| `BOR_LLM_EMBED_MODEL` | `embed` | Embedding model |
| `BOR_LLM_SUMMARY_MODEL` | `lite` | One-shot completions: document summaries at import, KB overview |
| `BOR_EMBEDDING_DIM` | `768` | Vector dimension (fixed at table creation) |
| `BOR_TOP_N_DOCS` | `2` | Full documents fed to the LLM |
| `BOR_RELEVANCE_THRESHOLD` | `0.62` | Answer when best cosine ≥ this **or** an FTS hit; below + no FTS ⇒ honest deflection |
| `BOR_HYBRID_VECTOR_CANDIDATES` | `100` | Cosine list width for the RRF fusion |
| `BOR_HYBRID_LEXICAL_CANDIDATES` | `30` | FTS list width for the RRF fusion |
| `BOR_RRF_K` | `60` | RRF damping constant (`1/(k + rank)`) |
| `BOR_AGENT_MAX_ROUNDS` | `10` | Hard cap on agent tool rounds per grounded turn (0 = no tools) |
| `BOR_IMPORT_EXTENSIONS` | csv (see below) | Importable formats (may only narrow the A9 set) |
| `BOR_GIT_SOURCES` | — (empty) | CSV of git repo URLs — **fallback while the admin Git sources page's list is empty** |
| `BOR_SOURCES_DIR` | `~/bor-sources` | Where git repos are cloned/pulled |
| `BOR_UPLOAD_DIR` | `~/bor-sources/uploads` | Where uploaded source archives are unpacked |
| `BOR_UPLOAD_MAX_MB` | `512` | Cap (MiB) for uploaded source archives (zip-bomb guard) |
| `BOR_STEERING_MAX_CHARS` | `8000` | Char budget for the `<tuning>` prompt section |
| `BOR_SUMMARY_MAX_CHARS` | `12000` | Cap on document content sent to the `lite` summary model |
| `BOR_KB_OVERVIEW_MAX_CHARS` | `4000` | Char budget for the `<knowledge_base>` prompt section |
| `BOR_OVERVIEW_INPUT_MAX_CHARS` | `40000` | Cap on the document list sent to `lite` for KB overview generation |
| `BOR_SUGGESTIONS` | built-in list | JSON seed for onboarding chips (shown before the first saved question) |
| `BOR_ADMIN_PASSWORD` | *(required)* | The single admin's password — app refuses to start when empty |
| `BOR_SESSION_SECRET` | *(required)* | Signing key for the `bor_session` cookie |
| `BOR_SESSION_MAX_AGE` | `43200` | Session-cookie lifetime in seconds (12 h, sliding) |
| `DEBUGPY` | `0` | `1` ⇒ attach-on-demand debugpy on `DEBUGPY_PORT` (default 5678) |
| `BOR_LOG_LEVEL` | `INFO` | App log level |

### Customizing the look

Every identity string is an env var: `BOR_APP_NAME` (display name),
`BOR_INPUT_PLACEHOLDER` (chat composer placeholder), and
`BOR_FOOTER_TEXT` (footer line) — all served by `GET /api/config` and
applied by `assets/brand.js` at boot.

The colors are set from the admin **Theme tab** (`/theme.html`,
admin-only — phase 91): the 8 identity colors plus the three strings
above are edited with text fields and color pickers, persisted in the
`ui_settings` table, and injected into every served page as an inline
`<style>` BEFORE first paint (no red flash, no pop-in). The old
CSS-file theming is retired — a leftover `BOR_THEME` line in a
deployment's `.env` is simply ignored. Leave everything unset and the
app renders the built-in dark-tech palette byte-identically.

---

## Checking retrieval quality

Ask the *real* pipeline (live aipi embeddings + the current KB) whether a
question lands on the right document, with gate verdict and per-document
cosine / FTS / fused scores:

```bash
uv run python -m scripts.eval_retrieval "How did I install gitlab?"
uv run python -m scripts.eval_retrieval --from-file questions.txt --top 8
```

Requires `AIPI_KEY` in the environment and an imported knowledge base.

---

## Debugging

`debugpy` is **off by default** and *never imported* unless you opt in:

```bash
DEBUGPY=1 uv run uvicorn app.main:app
# → log line: debugpy: remote debugging ENABLED, listening on 0.0.0.0:5678
```

Attach from VS Code (`.vscode/launch.json`):
```json
{
  "name": "Attach to Brain of Reese",
  "type": "debugpy",
  "request": "attach",
  "connect": { "host": "localhost", "port": 5678 },
  "pathMappings": [
    { "localRoot": "${workspaceFolder}", "remoteRoot": "/app" }
  ]
}
```

The port is non-blocking and attach-on-demand: the app keeps running
normally until you attach. Override the port with `DEBUGPY_PORT`.

---

## QA / Testing

Three layers — the project rule is **one story, one phase, one Playwright
suite** (see `AGENTS.md`):

```bash
# Unit + integration (FastAPI TestClient)
uv run pytest

# Coverage gate (phases require >90% on app/)
uv run pytest --cov=app --cov-report=term-missing

# Lint + static types
uv run ruff check .
uv run pyright

# Playwright E2E — install the browser once:
uv run playwright install chromium

# Each story's E2E runs IN ISOLATION:
uv run pytest tests/e2e/test_import_documents.py -v --no-cov
uv run pytest tests/e2e/test_chat_rag.py -v --no-cov
# ...one file per story in .agents/user_stories/ (see .agents/phases/todo/)
```

**Deterministic E2E:** by default the E2E app talks to a local **mock
aipi** (`tests/e2e/mock_llm.py`) whose embeddings are real token-overlap
vectors — so the cosine relevance threshold behaves like production. To
run E2E against the **live** self-hosted models instead:

```bash
E2E_REAL_LLM=1 uv run pytest tests/e2e/test_chat_rag.py -v --no-cov
```

---

## Production Deployment

Build the multi-stage image (frontend minified by esbuild in the builder
stage, deps installed by `uv`, non-root runtime):

```bash
podman build -t brain-of-reese/app:latest .
```

Run standalone (bring your own Postgres + pgvector):
```bash
podman run -d --name brain-of-reese \
  -p 8000:8000 \
  -e BOR_DATABASE_URL=postgresql+psycopg://reese:SECRETPASSWORD@dbhost:5432/brain_of_reese \
  -e BOR_LLM_BASE_URL=https://aipi.reeseapps.com/v1 \
  -e BOR_LLM_API_KEY=$AIPI_KEY \
  brain-of-reese/app:latest
```

The entrypoint runs `alembic upgrade head` automatically on start.

Or run the whole stack from compose (app + db):
```bash
podman compose --profile prod up -d --build
```

Production hardening: app runs as non-root (uid 10001), slim image,
healthcheck on `/api/health`, debugpy off unless `DEBUGPY=1`, all assets
served locally (no CDN), `BOR_ENVIRONMENT=production`.

---

## Troubleshooting

- **`401` from aipi** — set `BOR_LLM_API_KEY` (or `$AIPI_KEY`).
- **`litellm.UnsupportedParamsError … encoding_format`** — the aipi proxy
  (litellm `openai_like`) rejects the `encoding_format` parameter. The app
  already works around this by POSTing a minimal `{model, input}` payload.
  If you see this, you are likely calling the endpoint with a different
  client — drop the parameter (or set `litellm.drop_params = True` on the proxy).
- **Embedding dimension mismatch** — aipi changed models; run
  `uv run python -m scripts.llm_probe`, update `BOR_EMBEDDING_DIM`, then
  drop + recreate the chunks table.
- **Honest deflection** (the amber *"I haven't done anything like that"*
  bubble) — every question passes the honesty gate: deflection happens only
  when the best cosine similarity is below `BOR_RELEVANCE_THRESHOLD`
  (default `0.62`) **and** no chunk matched lexically (`fts_hits = 0`).
  A weak cosine with a lexical hit (name-your-tool questions) still gets a
  grounded answer. When it does deflect, the LLM prompt carries weak-hit
  *titles only* (no document content), the reply opens with *"I haven't
  done anything like that"*, the bubble renders amber with *"Maybe try"*
  chips derived from the closest indexed titles, and the `query_log` row
  records `deflected=true`. This is a feature, not a bug.
- **Answers deflect too often / too rarely** — tune
  `BOR_RELEVANCE_THRESHOLD` (lower = answers more, higher = more honest
  deflection): `0.0` ⇒ the gate leans entirely on FTS hits; `1.0` ⇒
  everything deflects unless a chunk matches lexically. The `embed` model's
  cosines cluster in a ~0.6–0.85 band on the live KB, so the default is
  `0.62`. Check real scores:
  ```sql
  SELECT question, top_score, fts_hits, deflected
  FROM query_log ORDER BY created_at DESC LIMIT 20;
  ```
- **KB offline banner in the chat** — Postgres isn't running:
  `podman compose up -d db`.
- **Stuck "Thinking…"** — the LLM is slow or down; a 120s client timeout
  turns it into an error banner automatically.

---

## Planning & Architecture

The architecture, LOCKED decisions, and the phase roadmap live in
[`.agents/PLAN.md`](.agents/PLAN.md); per-story specs in
[`.agents/user_stories/`](.agents/user_stories/).