* aidan] bug: fixed 2 auth claude login issue
* [aidan] bug: claude login w email+pw
* [aidan] ui: edited message for cases where anthropic redirects to enter email code even after login'
* [aidan] ui/ux: fix connecting status accuracy
Add agentBridge.ts and import it FIRST in index.tsx so the bridge is
installed on the window before any app code runs.
The visitor / door / doorbell analogy:
- The visitor is the agent. It shows up wanting to operate the app:
read the rules, see the controls, take an action, check what happened.
- The door is window.OPENSWARM_APP. A single, always-present surface the
visitor can knock on: describe() to learn the app, getState() to read a
snapshot, invoke(name, args) to act.
- The doorbell is register({ rules, controls, getState, invoke }). The app
presses it on mount to say "I'm home, here's how to talk to me." Until it
does, describe()/getState() answer { __ready: false } so the visitor
knows the app is still booting and waits, instead of concluding nobody
lives here.
Why a hook and not just an .md:
A markdown instruction ("please expose your controls on window.X") is a
request the app can forget, half-implement, or drift from, and nothing
fails when it does. The bridge ships WITH the template and is imported
before app code, so the door exists from first paint whether or not the
author thought about agents. The app's only job is to press the doorbell
once; it never wires the plumbing, so it cannot get the plumbing wrong.
The contract is executable, not aspirational, which is the difference
between every generated app being agent-operable by default and hoping
each one remembered to be.
An npm install killed mid-warm leaves node_modules/ populated but no
.bin/, so _ensure_warm_cache (which only checked the dir exists) trusted
the half-tree forever and every app symlinked to it died with
`vite: command not found`. Gate the cache on .bin/vite existing, and
harden the app's frontend/run.sh skip-install guard the same way.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Browser sub-agents can now drive Google Drive’s share-doc flow end to end: same-process cross-origin iframes (the share dialog) are visible through a frame-tree-aware AX walk, and combobox autocomplete fields fill in one click_index call via DOM.focus + Input.insertText. When a spawned browser card’s agent finishes, the card now fades through a “Task done” pill with a Keep button for ~3s instead of snapping out, so the user sees what happened.
Drop the inline 'Memory N% full' pill from AgentChat and the mode label from AgentCard so the chat header stays quiet until tools are connected. Backend now emits agent:context_update after compact and clear so the session token counters reflect the trimmed history instead of stale pre-compact totals, with a new _estimate_post_compact_input helper and matching invariants in test_v2_invariants.
* [aidan] ui: sidebar reorder
* [aidan] ux: sidebar naming bug fix
* [aidan] ui: title changing improvements
* [aidan] ui/ux: aux title gen streams for cx/ route + agent writes meta.json first
- All aux LLM calls (chat title, turn label, group meta, dashboard name) now stream
instead of using messages.create; 9router's cx/ subscription response translator
drops content for GPT-5-family models on the non-streaming path but works per-event.
- aux_max_tokens_for floors GPT-5 budget at 4K so reasoning still leaves headroom for
a label; non-reasoning models get the base 100.
- App Builder skill: write meta.json FIRST (step 1 of Quick start) so the app's name
surfaces in the sidebar + ViewEditor header on the agent's first tool call instead
of waiting until end-of-turn.
- Session-end sync_output_from_meta_json takes a fallback_name (= session.name) so an
agent that never writes meta.json still leaves the app with the aux-LLM chat title
rather than "Untitled App".
- Title display: truncateForTitle caps at 4 words / 30 chars; displayChatTitle picks
the right Phase 1 placeholder by session.mode; normalizeSessionName strips legacy
Agent-XXXX names at slice intake.
- Typewriter component (char-by-char delete-then-type) drives the chat header,
dashboard card title, sidebar Apps entry, and ViewEditor TextField + description
field; honors useReducedMotion.
* [aidan] tune: reduce gpt-5 aux token floor
duplicate_dashboard cloned the layout but left every browser_card pointing at
the source's browser_id and never copied the sessions, so the new dashboard
showed the browser but no chat. Now we mint fresh browser_ids, duplicate every
session attached to the source dashboard (in-memory + on-disk), remap
browser_id / parent_session_id / spawned_by / cards / expanded_session_ids onto
the new ids, and persist via _save_session. Also restores top-level agent cards
(layout.cards), which were silently dropped on duplicate.
When a user selects an App card on the dashboard via the "Select UI element" picker, thread the selected Output ids to the backend (mirroring the selected_browser_ids path) and inject a per-app context block into the system prompt: each app's absolute workspace path, entry point, meta.json, and a pointer to its SKILL.md. This lets the agent edit existing apps in place from the main view (the dashboard card's Vite runtime live-reloads on save) without switching to the Apps tab.
ciregenz