ur-agent 1.33.0 → 1.35.0

package/CHANGELOG.md

@@ -1,5 +1,40 @@
 # Changelog
 
+## 1.35.0
+
+- New `ur connect` CLI command (same implementation as the `/connect` slash
+  command): `ur connect status`, `ur connect <provider>`,
+  `ur connect <provider> --key <KEY>`, and `ur connect logout <provider>`.
+  Provider doctor fix-it hints that referenced `ur connect` now work as shown.
+- `ur spec run`/`ur spec verify` now accept the documented `--kernel` flag
+  from the CLI (previously only the slash command parsed it).
+- Hidden non-functional legacy commands from `--help`: `ur setup-token`,
+  `ur auth login`, `ur auth logout`, and the native-build `ur install`
+  (no native package is published; use `ur update`).
+- Removed dead external-bridge gating code left over from pre-1.34 behavior
+  (`UR_ENABLE_EXTERNAL_APP_PROVIDERS` and the persisted opt-in list); the env
+  var was already ignored at runtime.
+- Documentation overhaul: provider/model docs now match the 1.34 first-class
+  subscription-CLI behavior everywhere; new `docs/TROUBLESHOOTING.md`;
+  README command table covers the full public CLI; static docs site updated
+  (stale `install`/`setup-token` entries replaced with `connect`, `ide`,
+  `skill`, `task`, `sandbox`, `memory`, `local-first`, `update`).
+- Removed stale duplicate docs (`docs/AGENT_UPGRADE_1.15.0.md` …
+  `1.22.0.md`, `docs/CODE_FEATURE_INVENTORY.md`); release history lives in
+  this changelog.
+- Fixed `.gitignore` ignoring itself, so ignore rules ship with the repo.
+
+## 1.34.0
+
+- Restore the 1.30.3 subscription approach: Codex CLI, Claude Code, Gemini CLI
+  and Antigravity are first-class in `/model` again — shown by default and
+  usable directly (no `UR_ENABLE_EXTERNAL_APP_PROVIDERS` opt-in and no runtime
+  block). They dispatch through the official CLI; log in with
+  `ur auth <provider>`. The internal generic `subscription` placeholder is
+  hidden from listings.
+- API and local/server providers are unchanged: live model discovery from each
+  provider's `/models` endpoint and in-app masked API-key entry.
+
 ## 1.33.0
 
 - Add API keys from inside UR while it is running: in `/model`, selecting an

package/QUALITY.md

@@ -40,7 +40,7 @@ do not tag or publish until that workflow is green.
 Public feature releases should update the full documentation set:
 
 - root README
-- changelog and upgrade notes
+- `CHANGELOG.md`
 - `docs/`
 - static site files under `documentation/`
 - examples
@@ -49,10 +49,15 @@ Public feature releases should update the full documentation set:
 ## Runtime Assumptions
 
 - UR runs through Bun.
-- Model requests go to the local Ollama app at `http://localhost:11434/api`.
-- The local Ollama app may expose local models or Ollama Cloud-backed models.
-- UR does not call model provider APIs directly and does not manage model API
-  keys.
+- The default provider is the local Ollama app at `http://localhost:11434/api`;
+  it may expose local models or Ollama Cloud-backed models.
+- Selecting an API provider (OpenAI, Anthropic, Gemini, OpenRouter) makes UR
+  call that provider's API directly, using a key stored via `ur connect`
+  (OS keychain) or read from an environment variable. Keys are never written
+  to settings files.
+- Selecting a subscription CLI provider (Codex CLI, Claude Code, Gemini CLI,
+  Antigravity) dispatches turns through the vendor's official CLI. There is no
+  silent cross-provider fallback.
 - The GitHub install path runs `dist/cli.js`, so the bundle must match the
   package version.
 

package/README.md

@@ -69,8 +69,10 @@ handing work off to other tools or agents when needed.
 ### Requirements
 
 - Bun. This repository is configured with `bun@1.3.14`.
-- A Node.js-compatible shell environment.
-- A running Ollama app or server. UR defaults to `http://localhost:11434/api`.
+- A Node.js-compatible shell environment (Node 18+ for the npm launcher).
+- For the default local setup: a running Ollama app or server
+  (`http://localhost:11434/api`). Any other supported provider (API key,
+  OpenAI-compatible server, or subscription CLI) works without Ollama.
 - Optional tools for specific workflows: GitHub CLI, tmux, and supported IDE
   integrations.
 
@@ -157,12 +159,14 @@ environment variables. API, local, and OpenAI-compatible server providers are
 UR-native runtimes and behave like Ollama: UR owns the conversation loop, tool
 loop, errors, and output.
 
-The default provider list includes a generic `subscription` access entry so the
-access type is visible, but this build does not invent subscription models or
-route through provider apps. If no independent subscription runtime is
-configured, that entry is marked unavailable. Subscription CLI integrations are
-external app bridges; they are diagnostics/opt-in only and are disabled for
-normal runtime selection unless `UR_ENABLE_EXTERNAL_APP_PROVIDERS=1` is set.
+The provider list shows the subscription CLIs (Codex CLI, Claude Code, Gemini
+CLI, Antigravity) alongside the API and local/server providers, and they are
+first-class: pick one in `/model` and it dispatches through the official CLI.
+These are external app bridges — they run the vendor's own CLI using your
+subscription — so you log in with `ur auth <provider>` (e.g. `ur auth chatgpt`,
+`ur auth claude`). UR does not invent subscription models; each subscription
+shows its curated model list. The generic `subscription` entry is an internal
+placeholder and is hidden from the list.
 
 ```sh
 ur provider list
@@ -180,23 +184,33 @@ ur config set provider.fallback ollama
 
 Provider config accepts canonical IDs and common aliases. Examples:
 `openai-api`, `anthropic-api`, `gemini-api`, `openrouter`, `ollama`,
-`lmstudio`, `LM Studio`, `llama.cpp`, and `vllm`. External app bridge aliases
-such as `codex-cli`, `claude-code-cli`, `gemini-cli`, and `antigravity-cli` are
-accepted only when the external bridge opt-in is enabled.
-Use quotes for shell values with spaces.
+`lmstudio`, `LM Studio`, `llama.cpp`, `vllm`, and the subscription CLIs
+`codex-cli` (`chatgpt`), `claude-code-cli` (`claude`), `gemini-cli` (`gemini`),
+and `antigravity-cli` (`agy`). Use quotes for shell values with spaces.
+
+Connect accounts once with `ur connect` (or `/connect` in a session):
+
+```sh
+ur connect status                     # connection state for every provider
+ur connect codex-cli                  # subscription: launches the official login
+echo "$OPENAI_API_KEY" | ur connect openai-api   # API: store a key in the OS keychain
+ur connect logout openai-api          # clear a stored key
+```
 
 | Provider | Access type | Runtime kind | Legal path |
 | --- | --- | --- | --- |
-| Subscription | subscription | unavailable until configured | independent subscription runtime only |
-| OpenAI API | API key | UR-native | `OPENAI_API_KEY` |
-| Claude API | API key | UR-native | `ANTHROPIC_API_KEY` |
-| Gemini API | API key | UR-native | `GEMINI_API_KEY` |
-| OpenRouter | API/router | UR-native | `OPENROUTER_API_KEY` |
+| OpenAI API | API key | UR-native | `OPENAI_API_KEY` or `ur connect openai-api` |
+| Claude API | API key | UR-native | `ANTHROPIC_API_KEY` or `ur connect anthropic-api` |
+| Gemini API | API key | UR-native | `GEMINI_API_KEY` or `ur connect gemini-api` |
+| OpenRouter | API/router | UR-native | `OPENROUTER_API_KEY` or `ur connect openrouter` |
 | Ollama | local | UR-native | localhost Ollama runtime |
 | LM Studio | local/server | UR-native | local OpenAI-compatible server |
 | llama.cpp | local/server | UR-native | local OpenAI-compatible server |
 | vLLM | local/server | UR-native | OpenAI-compatible server |
-| External app bridges | subscription | opt-in diagnostics | disabled unless explicitly enabled |
+| Codex CLI | subscription | external app bridge | official Codex CLI login (`ur auth chatgpt`) |
+| Claude Code | subscription | external app bridge | official Claude Code login (`ur auth claude`) |
+| Gemini CLI | subscription | external app bridge | official Gemini Code Assist login (`ur auth gemini`) |
+| Antigravity | subscription | external app bridge | official Antigravity CLI login (`ur auth antigravity`) |
 
 #### Provider-first model selection
 
@@ -210,9 +224,10 @@ In the interactive app, `/model` is a two-step, provider-first picker:
    by source: `live` (discovered from the endpoint), `cache` (last discovery),
    or `static` (predefined). Local/server providers (Ollama, LM Studio,
    llama.cpp, vLLM) and OpenAI-compatible endpoints are discovered live; API
-   providers use their curated model list. The generic `subscription` entry has
-   no models unless a real independent subscription backend is configured.
-   External app bridges are hidden unless explicitly opted in.
+   providers use live discovery from their `/models` endpoint once a key is
+   connected (with a curated fallback list before that). Subscription CLIs show
+   their curated model list because the official CLIs expose no models API. The
+   generic `subscription` entry is an internal placeholder hidden from listings.
 
 Model lists never cross providers: OpenAI API, Claude API, Gemini API,
 OpenRouter, Ollama, and OpenAI-compatible local/server endpoints are separate
@@ -235,9 +250,9 @@ identity line in the system prompt reflects it too:
   OpenRouter on its OpenAI-compatible chat endpoint.
 - **Local/server** providers call the configured endpoint (`/v1/chat/completions`
   for LM Studio/llama.cpp/vLLM; the native API for Ollama).
-- **External app bridges** for subscription CLIs are disabled by default because
-  they delegate the turn to another agent app. Opt in with
-  `UR_ENABLE_EXTERNAL_APP_PROVIDERS=1` only when that behavior is intentional.
+- **Subscription CLIs** (Codex, Claude Code, Gemini, Antigravity) are external
+  app bridges: selecting one dispatches the turn through the vendor's official
+  CLI using your subscription. Log in with `ur auth <provider>`.
 - **Subscription** access does not list fake models. If no independent
   subscription backend is configured, `/model` marks it unavailable and asks you
   to choose a connected local, server, or API provider.
@@ -267,10 +282,12 @@ as first-class subcommands in the shipped CLI.
 | `ur ci-loop` | Run a build or test command, hand failures to a fix agent, and retry with a bounded budget. |
 | `ur test-first` | Detect the project stack, run compile/test/lint commands, store failure traces, and install edit-time verify gates. |
 | `ur safety` | Inspect or initialize project shell safety policy and evaluate command risk before execution. |
+| `ur sandbox` | Inspect and manage the sandbox/permission architecture: status, dependency check, policy init, and command approval levels. |
 | `ur context-pack` | Write project architecture context, task memory, and compressed context under `.ur/`. Supports memory kinds `decision`, `constraint`, `command`, `diff`, `note`, `architecture`, `preference`, `attempt`, `accepted`, and `rejected`. |
 | `/hooks` | Inspect lifecycle hooks (`BeforeEdit`, `AfterEdit`, `BeforeCommand`, `AfterCommand`, `BeforeCommit`, `OnFailure`) configured via settings files. |
 | `ur bg` | Run and manage detached local background agents with optional worktrees and PR creation. |
 | `ur worktree` | List, inspect, and clean up UR agent worktrees. |
+| `ur task` | Start, run, and hand off worktree-per-task sessions with optional PR creation. |
 | `ur automation` | Store and run project-local scheduled automation specs under `.ur/automations/`. |
 | `ur workflow` | Define, validate, graph, run, and resume declarative agent workflows. |
 | `ur crew` | Run a lead and worker subagent crew over a shared task board; `--decompose` auto-splits tasks with risk/tests/rollback metadata. |
@@ -279,6 +296,7 @@ as first-class subcommands in the shipped CLI.
 | `ur repo-edit` | Build a repo edit index, plan AST-aware renames, preview patches, and apply with rollback. |
 | `ur code-index` | Build, query, or watch a local semantic code index using Ollama embeddings; `ur code-index repo` adds files/symbols/calls/tests/docs/configs. |
 | `ur semantic-memory` | Build and search a project-local memory index. |
+| `ur memory retention` | Configure and apply local memory retention policies (TTL, max entries, decay). |
 | `ur knowledge` | Manage a curated project knowledge base with provenance. |
 | `ur artifacts` | Capture diffs, test runs, notes, and review feedback under `.ur/artifacts/`. |
 | `ur claim-ledger` | Map generated claims to file, web, MCP, tool, or user sources. |
@@ -289,7 +307,10 @@ as first-class subcommands in the shipped CLI.
 | `ur eval bench` | Import local SWE-bench, Terminal-Bench, or Aider Polyglot exports. |
 | `ur model-doctor` | Inspect Ollama models and report likely agent capabilities. |
 | `ur model-route` | Recommend a local model for a task by capability fit. |
+| `ur route` | Classify a task and recommend the best subagent and collaboration pattern. |
+| `ur local-first` | Show readiness for no-cloud, private, lab, offline, and edge/server environments. |
 | `ur provider` | List, check, and diagnose legal model provider adapters. |
+| `ur connect` | Connect a provider account: subscription login or store an API key in the OS keychain. |
 | `ur auth chatgpt` | Launch the official Codex CLI login for ChatGPT subscription access. |
 | `ur auth claude` | Launch the official Claude Code login flow. |
 | `ur auth gemini` | Use the official Gemini CLI login flow where supported. |
@@ -304,6 +325,15 @@ as first-class subcommands in the shipped CLI.
 | `ur a2a card` | Print UR-AGENT Card metadata for agent interoperability. |
 | `ur a2a serve` | Start an opt-in local A2A task server with bearer or delegation auth. |
 | `ur sdk` | Show programmatic headless usage and scaffold SDK examples. |
+| `ur trigger` | Parse a GitHub/Slack webhook payload and optionally launch a headless UR run. |
+| `ur agent-templates` | List or install reusable project agent templates. |
+| `ur agent-task` | Summarize task state, git diff status, and PR handoff commands. |
+| `ur agent-inspect` | Reconstruct a per-subagent timeline from a session transcript. |
+| `ur agent-features` | Show or initialize agent feature expansion scaffolds. |
+| `ur agent-trends` | Show UR coverage for current agent technology trends. |
+| `ur agents` | List configured agents. |
+| `ur doctor` | Check the health of the UR installation and auto-updater. |
+| `ur update` | Check npm for UR-AGENT updates (`ur upgrade` is an alias). |
 
 ### Status Bar
 
@@ -402,8 +432,8 @@ project-backed features: Agent surfaces (`ur`, `ur agents`, `ur crew`, `ur bg`),
 Rules (`AGENTS.md`, `UR.md`, `.cursor/rules/*.mdc`, `.cursorrules`, safety and
 guardrail config), MCP (`ur mcp`, `.mcp.json`, plugin MCP servers), Skills
 (`/skills`, bundled, project, user, and plugin skills), CLI (`ur --help`, `ur -p`,
-`ur exec`, `ur acp`), and Models (`ur model`, `ur model-doctor`, Ollama routing
-and discovery).
+`ur exec`, `ur acp`), and Models (`/model`, `ur config set model`,
+`ur model-doctor`, Ollama routing and discovery).
 
 Examples:
 
@@ -523,6 +553,20 @@ the permission boundary matters.
 See [Configuration](docs/CONFIGURATION.md), [Validation](docs/VALIDATION.md),
 and [Quality Notes](QUALITY.md) for operational guidance.
 
+## Troubleshooting
+
+Common problems — `ur` not found after install, a provider that is selected
+but not connected, a model that belongs to another provider, unreachable local
+servers, plugin or editor connection issues — are covered with symptom, cause,
+fix, and a verification command in the
+[Troubleshooting Guide](docs/TROUBLESHOOTING.md). Start with:
+
+```sh
+ur --version
+ur provider status
+ur provider doctor
+```
+
 ## Development
 
 Install dependencies:
@@ -560,24 +604,32 @@ The GitHub workflow runs production bundle, smoke, release, package, and global
 install checks only after the Bun test step succeeds. Do not publish or tag a
 release until that GitHub run is green.
 
+## Package
+
+- npm package: [`ur-agent`](https://www.npmjs.com/package/ur-agent), binary `ur`.
+- The published package ships the bundled CLI (`dist/cli.js`), launcher
+  (`bin/ur.js`), documentation (`docs/`, `documentation/`, `examples/`), and
+  first-party plugins (`plugins/`, `.ur-plugin` marketplace manifest is part of
+  the repository).
+- Releases follow the [Release Runbook](RELEASE.md); every release is recorded
+  in [CHANGELOG.md](CHANGELOG.md).
+
 ## Documentation
 
 - [Usage Guide](docs/USAGE.md)
 - [Configuration](docs/CONFIGURATION.md)
 - [Provider Guide](docs/providers.md)
+- [Troubleshooting](docs/TROUBLESHOOTING.md)
 - [Agent Feature Expansion](docs/AGENT_FEATURES.md)
 - [Agent Trend Coverage](docs/AGENT_TRENDS.md)
-- [1.22.0 Upgrade Notes](docs/AGENT_UPGRADE_1.22.0.md)
-- [1.20.0 Upgrade Notes](docs/AGENT_UPGRADE_1.20.0.md)
-- [1.19.0 Upgrade Notes](docs/AGENT_UPGRADE_1.19.0.md)
-- [1.18.0 Upgrade Notes](docs/AGENT_UPGRADE_1.18.0.md)
-- [1.17.0 Upgrade Notes](docs/AGENT_UPGRADE_1.17.0.md)
+- [Changelog](CHANGELOG.md)
 - [Development Guide](docs/DEVELOPMENT.md)
 - [IDE Guide](docs/IDE.md)
 - [ACP Guide](docs/ACP.md)
 - [Plugin Guide](docs/plugins.md)
 - [Validation Runbook](docs/VALIDATION.md)
 - [Release Runbook](RELEASE.md)
+- [Quality Notes](QUALITY.md)
 - [Contributing](CONTRIBUTING.md)
 - [Security](SECURITY.md)
 - [Static Documentation Site](documentation/index.html)

package/RELEASE.md

@@ -36,9 +36,19 @@ set and version:
 
 ```bash
 rg -n "Version [0-9]|expected: [0-9]|UR-AGENT v[0-9]" README.md docs documentation
-bun test test/docsCoverage.test.ts
+bun test test/docsCoverage.test.ts test/docsCommands.test.ts
 ```
 
+Version bump checklist (all three must move together):
+
+1. `package.json` `version`
+2. `bunfig.toml` `MACRO.VERSION`
+3. `documentation/index.html` version eyebrow
+4. `extensions/vscode-ur-inline-diffs/package.json` `version` (the VSIX test
+   requires it to match the root package version)
+5. Add a `CHANGELOG.md` entry, then run `bun run build` so `dist/cli.js`
+   embeds the new version (`bun run release:check` verifies all of this).
+
 If `npm whoami` fails, run:
 
 ```bash