ur-agent 1.83.2 → 1.84.1

package/docs/CONFIGURATION.md

@@ -182,15 +182,25 @@ entry. Its endpoint-scoped catalog is reused for five minutes, while Ctrl+R
 forces an immediate live refresh; a failed forced refresh never silently
 displays cached entries. API-provider secret
 entry stays on one masked row and stores the value through the keychain flow.
-Ollama and llama.cpp capabilities are loaded lazily for the focused model from
-`/api/show` and `/props`, respectively, so the arrow selector reflects the
-actual model rather than a provider-wide guess.
+Ollama, llama.cpp, and vLLM capabilities are loaded lazily for the focused
+model from `/api/show`, `/props`, and `/server_info?config_format=json`,
+respectively, so the arrow selector reflects live runtime evidence rather than
+a provider-wide guess. llama.cpp's current boolean support flag does not name
+accepted levels, while a configured vLLM reasoning parser exposes its
+documented `minimal→none|low|medium|high` mapping.
 
 The effort row contains only capability-backed selectors UR can map to native
 provider values. Ultra appears only when metadata advertises `ultra`, `max`,
 `xhigh`, or an explicit alias; mappings such as `ultra→max` are shown and sent
-exactly. Models that top out at `high`, boolean-thinking models, and unknown
-capabilities omit Ultra. See [Reasoning effort](providers.md#reasoning-effort).
+exactly. Models that top out at `high`, models without an advertised
+beyond-high value, and unknown capabilities omit Ultra. See
+[Reasoning effort](providers.md#reasoning-effort).
+
+Boolean-thinking models on runtimes with a native toggle expose a two-state
+control instead: Left selects off, Right selects on, and `t` toggles in
+`/model`; `/thinking on|off|status` is the direct command. This updates the live
+session and `alwaysThinkingEnabled`. Generic OpenAI-compatible transports do
+not receive an invented boolean field.
 
 The same provider-first picker is mandatory on the first interactive run in a
 workspace with no model in `.ur/settings.json` or `.ur/settings.local.json`.
@@ -719,6 +729,7 @@ Plugins can add commands, tools, and skills:
 
 ```sh
 ur plugin list
+ur plugin marketplace add npm:@scope/catalog@latest
 ur plugin search [query] [--capability <name>] [--marketplace <name>] [--installed] [--json]
 ur plugin show <name-or-name@marketplace> [--json]
 ur plugin install <plugin>

package/docs/USAGE.md

@@ -349,6 +349,10 @@ UR includes slash commands and CLI subcommands for common workflows:
   `ur plugin search [query]` for ranked cross-catalog discovery and
   `ur plugin show <name@marketplace>` to inspect provenance and capabilities
   before installation.
+  Add an npm-hosted catalog with
+  `ur plugin marketplace add npm:@scope/catalog@latest`; npm sources respect
+  the user's registry/authentication configuration and are refreshed with the
+  normal marketplace update command.
 - `ur agents` to list configured agents
 - `ur agent-trends` to inspect coverage for current agent technology trends
 - `ur a2a card` to print legacy Agent Card metadata, or

package/docs/VALIDATION.md

@@ -19,7 +19,7 @@ You need:
 
 ```sh
 ur --version
-# expected for this release: "1.83.2 (UR-Nexus)"
+# expected for this release: "1.84.1 (UR-Nexus)"
 ```
 
 ### 0.0 Redteam mode and Reverse Skills (1.81.0)
@@ -82,11 +82,18 @@ between models that top out at high, xhigh, max, and native-ultra models; the le
 selected ceiling must update immediately. Models that top out at high must omit Ultra, while
 xhigh/max entries must show `ultra→xhigh` or `ultra→max`, and the
 confirmation must match `/effort status` and the request wire value. For
-llama.cpp, verify focus requests
-`/props?model=<focused-id>` and that a template reporting
-`supports_reasoning_effort: false` has no graded selector. Open the OpenAI API or Claude
-API connection flow and verify the masked `API key` label and entry remain on
-one horizontal row.
+an Ollama model that advertises thinking without a model-specific ladder, verify that
+Left selects off, Right selects on, `t` toggles, and `/effort max` reports that
+max was not sent while enabling `think: true`; `/thinking off` must produce
+`think: false`. For
+llama.cpp, verify focus requests `/props?model=<focused-id>` and that both an
+unsupported template and a bare `supports_reasoning_effort: true` flag have no
+graded selector unless exact levels are also returned. For vLLM, verify one
+focus request to `/server_info?config_format=json`; a non-empty reasoning parser
+must expose `minimal→none`, `low`, `medium`, and `high`, serialize
+`minimal` as `reasoning_effort: "none"`, and omit Ultra. Open the OpenAI API or
+Claude API connection flow and verify the masked `API key` label and entry
+remain on one horizontal row.
 
 Then ask UR to research a current topic with WebSearch and WebFetch. Expected:
 the auxiliary request stays on the active OpenRouter model, no `modelH` error

package/docs/plugins.md

@@ -51,6 +51,50 @@ ur --plugin-dir ./plugins/community/my-plugin
 Plugins are loaded from local UR-Nexus paths first. Network marketplace installs
 remain explicit user actions and are subject to plugin policy checks.
 
+## Marketplace sources
+
+UR accepts GitHub shorthand, Git URLs, direct marketplace JSON URLs, npm
+packages, local files/directories, and inline settings manifests. Add an npm
+marketplace with the explicit `npm:` prefix:
+
+```sh
+ur plugin marketplace add npm:acme-ur-marketplace
+ur plugin marketplace add npm:@acme/ur-marketplace@latest
+ur plugin marketplace add npm:@acme/ur-marketplace@^2.0.0
+ur plugin marketplace update <marketplace-name>
+```
+
+The package must ship `.ur-plugin/marketplace.json`. An omitted version follows
+the registry's `latest` dist-tag; a version, semver range, or another dist-tag
+can be supplied after the package name, using npm's
+[package-spec syntax](https://docs.npmjs.com/cli/v11/using-npm/package-spec/).
+Refreshing the marketplace re-resolves
+that selector. UR uses the installed npm client, so standard `.npmrc`
+authentication, scoped registries, proxies, and registry settings continue to
+work. Package lifecycle scripts are disabled during marketplace download, and
+only the requested package—not its staging dependency tree—is retained.
+
+For a private registry selected in project or user settings:
+
+```json
+{
+  "extraKnownMarketplaces": {
+    "acme": {
+      "source": {
+        "source": "npm",
+        "package": "@acme/ur-marketplace",
+        "version": "^2.0.0",
+        "registry": "https://registry.example.com"
+      }
+    }
+  }
+}
+```
+
+After an install, removal, or external registry-file change,
+`/reload-plugins` clears both plugin discovery caches and the installed-plugin
+snapshot before reloading.
+
 ## Manifest reference
 
 A plugin is a directory containing `.ur-plugin/plugin.json`. UR uses a

package/docs/providers.md

@@ -182,31 +182,65 @@ For OpenRouter, UR preserves the live `/models` reasoning metadata and sends
 the unified `reasoning.effort` request. OpenAI-compatible servers receive the
 resolved value as `reasoning_effort`. The command confirmation, status
 indicator, active-work spinner, SDK settings response, and provider request all
-use the same resolved value. If a provider advertises only boolean thinking,
-UR does not invent a graded effort selector.
+use the same resolved value. If a provider advertises thinking without a
+model-specific graded ladder and its runtime has a real native on/off mapping,
+UR does not invent a graded effort
+selector. Use `/thinking on|off` directly;
+in `/model`, Left selects off, Right selects on, and `t` toggles. A graded
+`/effort` request on that model enables boolean thinking while clearly reporting
+that the requested level was not sent.
+Generic OpenAI-compatible endpoints have no universal boolean thinking field,
+so metadata alone does not make this toggle appear and UR sends no invented parameter.
 `ultra` is UR's visible beyond-high ceiling selector. It is selectable only
 when the provider/model advertises `ultra`, `max`, `xhigh`, or an explicit
 provider-authored equivalent. UR shows the native mapping (for example,
 `ultra→max`) and sends that exact wire value; it never enables Ultra for a model
-whose graded ladder tops out at `high`, boolean thinking, or unknown capability metadata. Arbitrary
+whose graded ladder tops out at `high`, lacks an advertised beyond-high value,
+or has unknown capability metadata. Arbitrary
 labels such as `deep` still require an explicit provider alias because UR
 cannot infer their rank.
 
+For an unknown or newly released model, UR waits for provider-authored model
+metadata or a supported model-scoped probe before adding thinking parameters.
+If the provider does not establish support, thinking stays off for request
+shaping; UR does not optimistically send an unknown parameter and treat an API
+error as capability discovery. Boolean thinking metadata enables the thinking
+toggle only and never invents a graded effort ladder. On OpenRouter, UR sends
+the provider-default `reasoning.enabled` control, or the exact token budget when
+the model advertises `supports_max_tokens`.
+
 For Ollama, UR lazily reads the focused model's `/api/show` capabilities and
-sends the selected level through native `think`. Kimi K3 uses
-`low|high|max` and therefore exposes Ultra as `ultra→max`; GPT-OSS uses
-`low|medium|high` and does not expose Ultra; other models advertising
-`thinking` use Ollama's current `low|medium|high|max` contract. Direct OpenAI,
+sends the resolved control through native `think`. A generic `thinking`
+capability proves thinking support but does not identify a model-specific
+graded ladder; UR therefore exposes the verified native on/off control without
+claiming that the model cannot also support levels. GPT-OSS uses Ollama's documented
+`low|medium|high` ladder and does not expose Ultra. Other graded ladders and
+Ultra aliases are used only when the endpoint explicitly returns them in model
+reasoning metadata.
+
+For vLLM, UR lazily reads the non-generating
+`/server_info?config_format=json` endpoint for the focused model. A configured
+reasoning parser establishes vLLM's documented Chat Completions contract:
+`none|low|medium|high`, displayed as `minimal→none|low|medium|high` and sent
+through `reasoning_effort`. This discovery never launches a completion and
+does not add Ultra. A richer provider-authored model record can add exact
+levels or aliases. For llama.cpp, `/props` can establish that the active chat
+template consumes reasoning effort, but the current capability flag does not
+publish its finite accepted values; UR does not fabricate a ladder from that
+boolean. Direct OpenAI,
 Anthropic, and Gemini models use curated model-specific ladders from their
 official documentation; live discovery rows are merged with those contracts.
 See [Ollama thinking](https://docs.ollama.com/capabilities/thinking),
 [OpenAI model guidance](https://developers.openai.com/api/docs/guides/latest-model),
 [Claude effort](https://platform.claude.com/docs/en/build-with-claude/effort),
-and [Gemini thinking](https://ai.google.dev/gemini-api/docs/thinking).
+[Gemini thinking](https://ai.google.dev/gemini-api/docs/thinking), and
+[vLLM reasoning outputs](https://docs.vllm.ai/en/latest/features/reasoning_outputs/).
 
 The provider-first `/model` picker supports the same control directly: use
-Left/Right to move through the capability-backed selectors UR can map to the focused model's native levels,
-then Enter to apply the model and effort together. OpenRouter's live catalog
+Left/Right to move through the capability-backed selectors UR can map to a
+graded model's native levels, or to choose off/on for a boolean-thinking model
+when its runtime has a native two-state mapping, then Enter to apply the model
+and reasoning control together. OpenRouter's live catalog
 shows pricing tier, context size, tool capability, reasoning capability, and
 the full, untruncated model ID immediately below the focused entry. Opening the
 OpenRouter catalog reuses its endpoint-scoped five-minute cache; Ctrl+R forces
@@ -218,13 +252,26 @@ provider prompt-cache markers. Explicit routing preferences and the `:nitro`,
 OpenAI, Claude, Gemini, and OpenRouter is a single aligned masked row; the key
 is stored in the OS keychain flow and is never written to settings.
 
+### Token counting
+
+UR uses each provider's non-generating count endpoint when one covers the full
+request: OpenAI Responses input tokens, Anthropic Messages token counting,
+Gemini `countTokens`, llama.cpp chat input tokens, and vLLM Messages token
+counting. Ollama, OpenRouter, LM Studio, Unsloth, and subscription CLIs use a
+provider-wire local estimate because those runtimes do not share a dependable
+preflight tokenizer for complete chat history plus tools. UR never launches a
+hidden completion for token counting. If a native count call is unavailable,
+file and MCP size checks retain the local estimate rather than disabling their
+limits.
+
 For llama.cpp, `/v1/models` metadata is preserved when the server supplies it.
-Because stock llama.cpp exposes chat-template effort support per loaded model,
 UR also resolves the model currently under the Up/Down cursor through
-`/props?model=<id>`. Left/Right is enabled only after that focused template
-advertises `supports_reasoning_effort`; the selected value is then sent
-unchanged in `reasoning_effort`. This works with llama.cpp router/cluster mode
-and does not assume that port 8080 limits UR to one worker.
+`/props?model=<id>`. `supports_reasoning_effort` establishes template support,
+but current llama.cpp does not expose the accepted value set through that flag,
+so it does not by itself enable Left/Right. Exact effort metadata from the
+model endpoint still enables the corresponding selectors and is sent unchanged
+as `reasoning_effort`. This works with llama.cpp router/cluster mode and does
+not assume that port 8080 limits UR to one worker.
 
 ### Provider-aware research calls
 

package/documentation/app.js

@@ -69,7 +69,7 @@ const featureGroups = [
     title: 'Providers and auth',
     tags: ['subscription', 'API', 'local', 'effort', 'status bar'],
     text: 'UR-native API/local/OpenAI-compatible runtimes, provider-scoped endpoints, provider-only Unsloth inference, capability-driven reasoning effort, responsive OpenRouter routing, first-class subscription CLI providers dispatched through the official vendor CLIs, provider doctor checks, secure API-key connect, non-secret config, fallback hints, and provider-aware status-bar output.',
-    commands: ['ur provider list', 'ur provider status', 'ur provider doctor agy', 'ur connect status', 'ur config set provider openai-api', 'ur config set provider ollama', 'ur config set base_url llama.cpp http://localhost:9931/v1', '/effort ultra'],
+    commands: ['ur provider list', 'ur provider status', 'ur provider doctor agy', 'ur connect status', 'ur config set provider openai-api', 'ur config set provider ollama', 'ur config set base_url llama.cpp http://localhost:9931/v1', '/effort ultra', '/thinking on'],
   },
   {
     title: 'Security and operations',
@@ -85,7 +85,7 @@ const commands = [
     category: 'Core',
     aliases: [],
     summary: 'Start an interactive session; a fresh workspace must choose and locally persist a validated provider/model pair first.',
-    examples: ['ur', 'ur --model kimi-k3:cloud --effort ultra', 'ur --continue', 'ur --resume'],
+    examples: ['ur', 'ur --model gpt-5.6-sol --effort ultra', 'ur --continue', 'ur --resume'],
   },
   {
     name: 'ur -p',
@@ -378,8 +378,8 @@ const commands = [
     name: 'plugin',
     category: 'Interop',
     aliases: ['plugins'],
-    summary: 'Manage UR plugins and marketplaces for MCP tools, skills, templates, validators, language adapters, LSP servers, agents, hooks, output styles, and commands.',
-    examples: ['ur plugin search git', 'ur plugin search --capability skills --json', 'ur plugin show github@ur-plugins-official', 'ur plugin list', 'ur plugin install hello@ur-plugins-official', 'ur plugin update <plugin>'],
+    summary: 'Manage UR plugins and GitHub, Git, URL, npm, local, or settings-backed marketplaces for MCP tools, skills, templates, validators, language adapters, LSP servers, agents, hooks, output styles, and commands.',
+    examples: ['ur plugin marketplace add npm:@scope/catalog@latest', 'ur plugin search git', 'ur plugin search --capability skills --json', 'ur plugin show github@ur-plugins-official', 'ur plugin list', 'ur plugin install hello@ur-plugins-official', 'ur plugin update <plugin>'],
   },
   {
     name: 'provider',
@@ -547,8 +547,8 @@ const slashGroups = [
   },
   {
     title: 'Models, tools, and interop',
-    items: ['/model', '/provider', '/effort', '/fast', '/model-doctor', '/model-route', '/escalate', '/mcp', '/plugin', '/skills', '/skill', '/sdk', '/a2a-card'],
-    text: 'Pick providers and models, cycle only capability-backed effort selectors that map to provider-native values, inspect capabilities, manage MCP/plugin extensions, browse prompt skills with /skills, run executable workflows with /skill, and expose interop surfaces.',
+    items: ['/model', '/provider', '/effort', '/thinking', '/fast', '/model-doctor', '/model-route', '/escalate', '/mcp', '/plugin', '/skills', '/skill', '/sdk', '/a2a-card'],
+    text: 'Pick providers and models, cycle only capability-backed effort selectors or a verified provider-native thinking toggle, inspect capabilities, manage MCP/plugin extensions, browse prompt skills with /skills, run executable workflows with /skill, and expose interop surfaces.',
   },
   {
     title: 'Security operations',

package/documentation/index.html

@@ -45,7 +45,7 @@
     <main id="content" class="content">
       <header class="topbar">
         <div>
-          <p class="eyebrow">Version 1.83.2</p>
+          <p class="eyebrow">Version 1.84.1</p>
           <h1>UR-Nexus Documentation</h1>
           <p class="lead">A practical, tutorial-style reference for installing, configuring, automating, extending, and operating UR-Nexus.</p>
         </div>
@@ -76,7 +76,7 @@
           </article>
           <article>
             <strong>Plugin marketplace</strong>
-            <span>Plugins can add MCP tools, skills, templates, validators, language adapters, LSP servers, agents, hooks, and output styles.</span>
+            <span>GitHub, Git, URL, npm, local, and settings-backed catalogs can add MCP tools, skills, templates, validators, language adapters, LSP servers, agents, hooks, and output styles.</span>
           </article>
           <article>
             <strong>Legal provider routing</strong>
@@ -100,7 +100,7 @@
           </article>
           <article>
             <strong>Deterministic commands</strong>
-            <span>The external runtime exposes 172 commands and 253 unique slash tokens; registry tests reject ambiguous names, broken loaders, and undocumented visible commands.</span>
+            <span>The external runtime exposes a source-derived command catalog; registry tests reject ambiguous names, broken loaders, undocumented visible aliases, and documented commands without implementations.</span>
           </article>
           <article>
             <strong>Safety and context</strong>
@@ -177,9 +177,10 @@ ur config set responses.store false</code></pre>
           <article>
             <h3>Capability-driven reasoning effort</h3>
             <pre><code>/effort ultra
+/thinking on
 ur --model kimi-k3:cloud --effort high
-/model  # Up/Down model · Left/Right effort · Enter apply</code></pre>
-            <p>The normalized vocabulary is <code>minimal</code>, <code>low</code>, <code>medium</code>, <code>high</code>, <code>xhigh</code>, <code>max</code>, and <code>ultra</code>; <code>/effort auto</code> clears an explicit choice. UR lists only capability-backed selectors it can map to the focused model's provider-native levels. <code>max</code> resolves to the highest supported non-Ultra tier. Ultra appears only for native <code>ultra</code>, advertised <code>max</code>/<code>xhigh</code>, or an explicit provider alias; the picker shows translations such as <code>ultra→max</code> and sends that exact provider value. Models that top out at <code>high</code>, boolean-thinking models, and unknown capabilities do not get Ultra.</p>
+/model  # Up/Down model · Left/Right effort or thinking on/off · Enter apply</code></pre>
+            <p>The normalized vocabulary is <code>minimal</code>, <code>low</code>, <code>medium</code>, <code>high</code>, <code>xhigh</code>, <code>max</code>, and <code>ultra</code>; <code>/effort auto</code> clears an explicit choice. UR lists only capability-backed selectors it can map to the focused model's provider-native levels. <code>max</code> resolves to the highest supported non-Ultra tier. Ultra appears only for native <code>ultra</code>, advertised <code>max</code>/<code>xhigh</code>, or an explicit provider alias; the picker shows translations such as <code>ultra→max</code> and sends that exact provider value. Models that top out at <code>high</code>, lack an advertised beyond-high value, or have unknown capabilities do not get Ultra. For models with thinking but no advertised graded ladder on runtimes with a native toggle, Left selects off, Right selects on, and <code>t</code> toggles; <code>/thinking on|off</code> is the direct control. A graded <code>/effort</code> request on such a model enables thinking while reporting that no graded value was sent. Focused vLLM models use non-generating <code>/server_info</code> discovery for the documented <code>minimal→none|low|medium|high</code> mapping; llama.cpp's bare support flag creates no invented ladder. Generic OpenAI-compatible runtimes receive no invented boolean field.</p>
           </article>
           <article>
             <h3>OpenRouter responsive routing</h3>
@@ -189,7 +190,7 @@ ur --model kimi-k3:cloud --effort high
           </article>
           <article>
             <h3>Status bar and updates</h3>
-            <pre><code>Ollama | llama3 | ask | main | update 1.83.2 available</code></pre>
+            <pre><code>Ollama | llama3 | ask | main | update available</code></pre>
             <p>The interactive status bar shows only important runtime state: provider, model, mode, branch, active tasks, checks status when known, and update availability. It is hidden in CI, dumb terminals, and print mode.</p>
           </article>
           <article>

package/extensions/jetbrains-ur/build.gradle.kts

@@ -7,7 +7,7 @@ plugins {
 }
 
 group = "dev.urnexus"
-version = "1.83.2"
+version = "1.84.1"
 
 repositories {
     mavenCentral()

package/extensions/vscode-ur-inline-diffs/package.json

@@ -2,7 +2,7 @@
   "name": "ur-inline-diffs",
   "displayName": "UR Inline Diffs",
   "description": "Review, apply, and reject UR inline diff bundles from .ur/ide/diffs inside VS Code.",
-  "version": "1.83.2",
+  "version": "1.84.1",
   "publisher": "ur-nexus",
   "engines": {
     "vscode": "^1.92.0"

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ur-agent",
-  "version": "1.83.2",
+  "version": "1.84.1",
   "description": "UR-Nexus — autonomous engineering workflow engine (plan, execute, test, verify, document, benchmark, reproduce)",
   "type": "module",
   "packageManager": "bun@1.3.14",