aiden-runtime 3.18.0 → 3.19.4

package/README.md

@@ -33,8 +33,18 @@ Self-Healing • Browser Automation • Terminal Control • Persistent Memory
 
 ---
 
-> **v3.18 — live dropdown UX · real PC control · smart model failover · anti-confabulation**
-> Type `/` for 63 commands or `@` for 61 tools with instant dropdown. Open/close apps, change volume, and control your PC for real — no more fake responses. Smart per-model failover with free-tier defaults. See [changelog](#changelog) below.
+> **v3.19.0 "ALIVE" — identity refreshes every turn · real-time state via tools · honest failure diagnostics · plugin slash commands**
+> Type `/` for 91 commands or `@` for 89 tools with instant dropdown. SOUL/USER/GOALS refresh every turn (not every 40 messages). Tool failures name the provider, retries, and fallback. Plugins extend slash commands via `commandCatalog.register()`. See [changelog](#changelog) below.
+
+---
+
+## Support Aiden
+
+Solo-built, AGPL-3.0. If Aiden saves you time, consider sponsoring development.
+
+[Sponsor →](https://razorpay.me/@whitelotus9625)
+
+Funds go to ongoing development, infrastructure costs, and contributor bounties.
 
 ---
 
@@ -48,6 +58,10 @@ Most AI agents answer questions. Aiden executes work.
 - **Learns from every session** — writes skills from successes, lessons from failures
 - **Works fully offline** — Ollama support, zero cloud dependency
 - **One command to start** — `npx aiden-os` installs, configures, runs everything
+- **Lives where you do** — identity (SOUL/USER/GOALS) refreshes every turn, not every 40 messages; edit `USER.md` mid-conversation and the change lands within one reply
+- **Real-time state via tools** — `now_playing` and friends query live system state instead of returning stale snapshots cached at startup
+- **Honest failures** — every tool failure names the tool, provider, retry count, fallback chain, error, and next step; no silent swallowing
+- **Plugin extension** — drop a `.js` file into `workspace/plugins/` and call `ctx.commandCatalog.register('/mycommand', …)` to add slash commands without touching core
 
 ---
 
@@ -166,6 +180,34 @@ Set `AIDEN_HEADLESS=true` to suppress the Electron GUI when running the packaged
 
 ---
 
+## Getting Started
+
+Once Aiden is running, type these in the chat prompt:
+
+| First thing to do | What to type |
+|---|---|
+| See all available commands | `/help` |
+| Check which AI provider is active | `/switch` |
+| See your daily token budget | `/budget` |
+| Browse available skills | `/skills` |
+| Install a skill from the registry | `/install <skill-name>` |
+| Open the web UI in a browser | navigate to `localhost:4200/ui` |
+| Check model availability | `/models` |
+
+**Ask anything in plain English** — no special syntax needed for regular tasks:
+
+```
+summarize the PDF on my desktop
+open chrome and search for latest AI news
+close spotify
+take a screenshot and describe what you see
+what files did I download today
+```
+
+Type `/` to browse all 91 commands with instant search. Type `@` to select any of 89 tools directly.
+
+---
+
 ## Troubleshooting
 
 **"Cannot find module" or TypeScript errors**
@@ -241,20 +283,15 @@ Full chat interface with live activity panel. Local-first, connects to Ollama or
 
 ---
 
-## Aiden vs alternatives
+## What Aiden is
 
-| Feature | Aiden | Hermes | OpenClaw |
-|---|---|---|---|
-| **Windows native installer** | ✅ | ❌ | ❌ |
-| **Desktop OS control** | ✅ vision + mouse + keyboard | ❌ | ❌ |
-| **One-command install** | ✅ `npx aiden-os` | ❌ | ❌ |
-| **OpenAI-compatible API** | ✅ `/v1/chat/completions` | ❌ | ❌ |
-| **agentskills.io skills** | ✅ 1500+ | ✅ | ✅ 13K+ |
-| **Offline (Ollama)** | ✅ | ✅ | ✅ |
-| **Local web dashboard** | ✅ | ✅ | ✅ |
-| **Pro license system** | ✅ | ❌ | ❌ |
-| **Zero CVEs** | ✅ | ❌ | ❌ |
-| **License** | AGPL-3.0 | MIT | MIT |
+Aiden runs locally on your machine. It controls your desktop, browser, and terminal through natural conversation. It learns from your work and remembers what matters across sessions.
+
+- **Local-first** — your conversations and data stay on your machine. No cloud account required.
+- **Real desktop control** — vision, mouse, keyboard, browser, terminal, files. Not a chatbot wrapped in a sandbox.
+- **Persistent memory** — Aiden remembers facts, preferences, and lessons from prior sessions. The longer you use it, the better it knows your work.
+- **Honest by design** — when a tool fails, Aiden surfaces the failure rather than fabricating success.
+- **Open source** — AGPL-3.0. Read every line, modify anything, contribute back.
 
 ---
 
@@ -424,10 +461,43 @@ Or toggle live from the Aiden CLI without restarting:
 
 ---
 
+## CLI vs Dashboard Quick Reference
+
+Both the terminal TUI and the browser dashboard (`localhost:4200/ui`) expose the full feature set. Use whichever fits your workflow.
+
+| Feature | Terminal CLI | Browser (`localhost:4200/ui`) |
+|---|---|---|
+| Chat | ✅ inline prompt | ✅ chat panel |
+| Streaming responses | ✅ token-by-token | ✅ live SSE |
+| Markdown rendering | ✅ | ✅ |
+| Slash commands | ✅ `/help`, `/switch`, `/budget`… | ✅ same commands |
+| `/` command dropdown | ✅ instant, 91 commands | ✅ |
+| `@` tool picker | ✅ instant, 89 tools | ✅ |
+| Provider panel | `/switch` | ✅ Providers tab |
+| Memory panel | `/memory` | ✅ Memory tab |
+| Skills panel | `/skills` | ✅ Skills tab |
+| Plugin hooks | ✅ | ✅ |
+| MCP server mode | `aiden mcp` | — |
+| OpenAI-compatible API | — | ✅ `localhost:4200/v1` |
+
+---
+
 ## Contributing
 
 Contributions are welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for the full guide.
 
+**Quickstart:**
+
+```bash
+git clone https://github.com/taracodlabs/aiden.git
+cd aiden
+npm install
+cp .env.example .env   # add at minimum one API key (Groq is free: console.groq.com/keys)
+npm run build
+npm start              # server on :4200
+npm run cli            # TUI in a second terminal
+```
+
 - Bug fixes and new skills are the easiest entry points
 - All contributors sign the [CLA](.github/CLA.md) once via PR comment
 - Follow [Conventional Commits](https://www.conventionalcommits.org/)
@@ -435,21 +505,85 @@ Contributions are welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for the ful
 
 ---
 
-## Resources
+## Community
 
 | | |
 |---|---|
-| **Download installer** | [Latest release](https://github.com/taracodlabs/aiden-releases/releases/latest) |
+| **Discord** | [discord.gg/gMZ3hUnQTm](https://discord.gg/gMZ3hUnQTm) — chat, support, share what you build |
+| **Skills registry** | [agentskills.io](https://agentskills.io) — 1,500+ community skills |
+| **Bug reports & features** | [github.com/taracodlabs/aiden/issues](https://github.com/taracodlabs/aiden/issues) |
+| **Star the repo** | [github.com/taracodlabs/aiden](https://github.com/taracodlabs/aiden) ⭐ |
+| **npm** | [`npm install -g aiden-os`](https://www.npmjs.com/package/aiden-os) |
+| **Sponsor** | [github.com/sponsors/taracodlabs](https://github.com/sponsors/taracodlabs) |
+
+---
+
+## Documentation
+
+| Document | Description |
+|---|---|
+| [ARCHITECTURE.md](ARCHITECTURE.md) | Layer-by-layer breakdown, data flow diagrams, skill system design |
+| [CONTRIBUTING.md](CONTRIBUTING.md) | How to contribute — skills, tools, providers, docs |
+| [docs/ROADMAP.md](docs/ROADMAP.md) | Planned features and milestone tracker |
+| [docs/mcp/](docs/mcp/) | MCP server setup — Claude Code, Cursor, VS Code integration |
+| [.env.example](.env.example) | All ~90 environment variables with descriptions |
+| [workspace-templates/](workspace-templates/) | Starter workspace configs and example plugins |
+| **Download installer** | [github.com/taracodlabs/aiden-releases/releases/latest](https://github.com/taracodlabs/aiden-releases/releases/latest) |
 | **Releases & changelog** | [github.com/taracodlabs/aiden-releases](https://github.com/taracodlabs/aiden-releases) |
 | **License** | AGPL-3.0 core · Apache-2.0 skills |
 
 ---
 
+## Migration from Other Agents
+
+> Short version (no breaking changes in v3.18 or v3.19 — existing plugins and `.env` files continue to work):
+
+- **Skills** — Aiden is fully compatible with [agentskills.io](https://agentskills.io). Any skill with a valid `skill.json` manifest loads automatically via `/install <name>`.
+- **API clients** — Aiden exposes an OpenAI-compatible API at `localhost:4200/v1`. If you pointed your client at another agent, update the base URL and you're done.
+- **Config / env** — Most standard keys (`OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GROQ_API_KEY`, etc.) are recognized as-is. Copy your existing `.env` and Aiden picks them up on first start.
+
+---
+
 ## Changelog
 
+### v3.19.0 — 2026-05-01 "ALIVE"
+
+**Source-of-truth tool registry**
+- 13 hand-maintained tool lists collapsed into a single `TOOL_REGISTRY`; every derived list is generated, not maintained
+- Validator throws on drift at startup — no more silent orphan tools
+- 12 previously-unreachable tools now reachable by the planner (`fetch_url`, `cmd`, `ps`, `wsl`, `git_status`, `manage_goals`, `get_calendar`, `read_email`, `send_email`, `ingest_youtube`, `schedule_reminder`, `compact_context`)
+
+**Per-turn protected context refresh**
+- SOUL / USER / GOALS / STANDING_ORDERS / LESSONS refresh every turn instead of every 40 messages
+- Hash-based file-level cache — 24× token reduction on stable turns (3,860 → 159 tokens)
+- Edit `USER.md` mid-conversation; Aiden picks up changes within one reply
+
+**Honesty enforcement**
+- 5 fake InstantActions removed (`screenshot`, `volume_up/down/mute`, `lock_screen`) — now route to real handlers, surface real errors
+- Action-verb planner guard: prevents respond-only plans for action intents
+- Diagnostic failure messages: every failure names tool, provider, retries, fallback, error, suggestion
+- Hidden bug fixed: `fastPath` was bypassing the planner for short action messages
+
+**Real-time state via tools**
+- `now_playing` tool: live media query via Windows `WinRT GlobalSystemMediaTransportControlsSessionManager`
+- Volatile startup state dump removed from system prompt
+- SOUL.md lazy-state rule: model calls tools for current state instead of using cached values
+
+**Registry-backed slash completer**
+- `commandCatalog` is the single source of truth for slash commands (91 total)
+- 6 previously-invisible commands now in dropdown (`/plugins`, `/profile`, `/failed`, `/install`, `/publish`, `/sandbox`)
+- Plugin contribution path live: plugins call `commandCatalog.register()` to add commands
+- Generation-cached dropdown: rebuild only when catalog changes
+
+**Provider chain expansion**
+- NVIDIA NIM promoted from executor-only to chat slots (`nvidia-1`, `nvidia-2`)
+- Chain order: `groq → gemini → nvidia → openrouter → together → ollama`
+
+---
+
 ### v3.18.0 — 2026-04-30
 
-**Live dropdown UX (Hermes-style)**
+**Live dropdown UX**
 - Type `/` for instant command dropdown (63 commands)
 - Type `@` for tool dropdown (61 tools)
 - Prefix-match filter, arrow nav, Tab to select, Esc to close
@@ -488,11 +622,6 @@ Contributions are welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for the ful
 **Skill loader fix**
 - 1,484 skills now load (was blocking 1,445 due to overly broad patterns)
 
-**Known issues — fixing in v3.19**
-- Cross-provider failover not always reliable (Groq may not try other providers)
-- Real-time state queries (now playing, open tabs) need dedicated tools
-- YouTube auto-click occasionally fails on slow-loading pages
-
 ---
 
 ### v3.17.0 — 2026-04-28
@@ -570,7 +699,7 @@ Contributions are welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) for the ful
 
 **Ecosystem & Interoperability**
 - OpenAI-compatible API — `/v1/chat/completions` + `/v1/models`. Point Open WebUI, LibreChat, Cursor, or any OpenAI SDK at `localhost:4200` and get Aiden's full 89-tool agent (not just raw LLM inference)
-- agentskills.io compatibility — skills now ship with `skill.json` manifest. Compatible with Hermes, OpenClaw, and any agentskills.io agent. 1,515 existing skills backfilled automatically
+- agentskills.io compatibility — skills now ship with `skill.json` manifest. Compatible with the agentskills.io specification. 1,515 existing skills backfilled automatically
 - Streaming tool output — shell commands, Python scripts, and browser extraction stream live progress lines as they execute. Set `AIDEN_SHOW_TOOL_OUTPUT=false` to suppress
 
 ---

package/config/devos.config.backup.json

@@ -0,0 +1,225 @@
+{
+  "user": {
+    "name": "shiva"
+  },
+  "model": {
+    "active": "groq-1",
+    "activeModel": "llama-3.3-70b-versatile"
+  },
+  "providers": {
+    "ollama": {
+      "enabled": true,
+      "models": []
+    },
+    "apis": [
+      {
+        "model": "gemini-3-flash",
+        "enabled": false,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "boa-1",
+        "provider": "boa",
+        "key": "boa-5a949b90-094c-4f35-a6ae-3fab8acc42aa"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": true,
+        "usageCount": 72,
+        "name": "groq-1",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY",
+        "rateLimitedAt": 1777579113083,
+        "rateLimitWindow": 300000
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": true,
+        "usageCount": 105,
+        "name": "groq-2",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_2"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": true,
+        "usageCount": 41,
+        "name": "groq-3",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_3"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": true,
+        "usageCount": 12,
+        "name": "groq-4",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_4"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": true,
+        "usageCount": 7,
+        "name": "groq-5",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_5"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "groq-6",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_6"
+      },
+      {
+        "model": "gemini-2.5-flash",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 17,
+        "name": "gemini-1",
+        "provider": "gemini",
+        "key": "env:GEMINI_API_KEY"
+      },
+      {
+        "model": "gemini-2.5-flash",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 9,
+        "name": "gemini-2",
+        "provider": "gemini",
+        "key": "env:GEMINI_API_KEY_2"
+      },
+      {
+        "model": "gemini-2.5-flash",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 9,
+        "name": "gemini-3",
+        "provider": "gemini",
+        "key": "env:GEMINI_API_KEY_3"
+      },
+      {
+        "model": "gemini-2.5-flash",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 9,
+        "name": "gemini-4",
+        "provider": "gemini",
+        "key": "env:GEMINI_API_KEY_4"
+      },
+      {
+        "model": "nvidia/llama-3.3-nemotron-super-49b-v1",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "nvidia-1",
+        "provider": "nvidia",
+        "key": "env:NVIDIA_API_KEY"
+      },
+      {
+        "model": "meta/llama-3.3-70b-instruct",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "nvidia-2",
+        "provider": "nvidia",
+        "key": "env:NVIDIA_API_KEY"
+      },
+      {
+        "model": "meta-llama/llama-3.3-70b-instruct:free",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 7,
+        "name": "openrouter-1",
+        "provider": "openrouter",
+        "key": "env:OPENROUTER_API_KEY"
+      },
+      {
+        "model": "meta-llama/llama-3.3-70b-instruct:free",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 6,
+        "name": "openrouter-2",
+        "provider": "openrouter",
+        "key": "env:OPENROUTER_API_KEY_2"
+      },
+      {
+        "model": "meta-llama/llama-3.3-70b-instruct:free",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 6,
+        "name": "openrouter-3",
+        "provider": "openrouter",
+        "key": "env:OPENROUTER_API_KEY_3"
+      },
+      {
+        "model": "meta-llama/llama-3.1-405b-instruct",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 7,
+        "name": "together-1",
+        "provider": "custom",
+        "baseUrl": "https://api.together.xyz/v1/chat/completions",
+        "key": "env:TOGETHER_API_KEY"
+      },
+      {
+        "model": "gpt-4o-mini",
+        "enabled": false,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "boa-2",
+        "provider": "boa",
+        "key": "boa-5a949b90-094c-4f35-a6ae-3fab8acc42aa"
+      }
+    ]
+  },
+  "onboardingComplete": true,
+  "telegram": {
+    "enabled": false,
+    "botToken": "",
+    "allowedChatIds": [],
+    "pollingInterval": 1000
+  },
+  "routing": {
+    "mode": "auto",
+    "fallbackToOllama": true
+  },
+  "ollama": {
+    "fallbackModels": [],
+    "baseUrl": "http://localhost:11434",
+    "model": "gemma4:e4b",
+    "plannerModel": "gemma4:e4b",
+    "coderModel": "qwen2.5-coder:7b",
+    "fastModel": "llama3.2:latest"
+  },
+  "customProviders": [
+    {
+      "id": "bayofassets-haiku",
+      "displayName": "BayOfAssets Haiku 4.5",
+      "baseUrl": "https://api.bayofassets.com/v1/chat/completions",
+      "apiKey": "boa-dd14fba0-cdee-4247-ba03-90de01674c06",
+      "model": "claude-haiku-4-5",
+      "enabled": false,
+      "tier": 1
+    },
+    {
+      "id": "bayofassets",
+      "displayName": "Bay of Assets",
+      "baseUrl": "https://api.bayofassets.com/v1",
+      "apiKey": "boa-5a949b90-094c-4f35-a6ae-3fab8acc42aa",
+      "model": "gemini-3-flash",
+      "enabled": false,
+      "tier": 2
+    }
+  ],
+  "cli": {
+    "theme": "ember"
+  },
+  "primaryProvider": ""
+}

package/config/devos.config.json

@@ -23,81 +23,127 @@
       },
       {
         "model": "llama-3.3-70b-versatile",
-        "enabled": true,
+        "enabled": false,
         "rateLimited": false,
-        "usageCount": 63,
+        "usageCount": 97,
         "name": "groq-1",
         "provider": "groq",
         "key": "env:GROQ_API_KEY"
       },
       {
         "model": "llama-3.3-70b-versatile",
-        "enabled": true,
-        "rateLimited": false,
-        "usageCount": 79,
+        "enabled": false,
+        "rateLimited": true,
+        "usageCount": 105,
         "name": "groq-2",
         "provider": "groq",
         "key": "env:GROQ_API_KEY_2"
       },
       {
         "model": "llama-3.3-70b-versatile",
-        "enabled": true,
-        "rateLimited": false,
-        "usageCount": 37,
+        "enabled": false,
+        "rateLimited": true,
+        "usageCount": 41,
         "name": "groq-3",
         "provider": "groq",
         "key": "env:GROQ_API_KEY_3"
       },
       {
         "model": "llama-3.3-70b-versatile",
-        "enabled": true,
-        "rateLimited": false,
-        "usageCount": 11,
+        "enabled": false,
+        "rateLimited": true,
+        "usageCount": 12,
         "name": "groq-4",
         "provider": "groq",
         "key": "env:GROQ_API_KEY_4"
       },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": false,
+        "rateLimited": true,
+        "usageCount": 7,
+        "name": "groq-5",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_5"
+      },
+      {
+        "model": "llama-3.3-70b-versatile",
+        "enabled": false,
+        "rateLimited": false,
+        "usageCount": 6,
+        "name": "groq-6",
+        "provider": "groq",
+        "key": "env:GROQ_API_KEY_6"
+      },
       {
         "model": "gemini-2.5-flash",
-        "enabled": true,
+        "enabled": false,
         "rateLimited": false,
-        "usageCount": 16,
+        "usageCount": 26,
         "name": "gemini-1",
         "provider": "gemini",
         "key": "env:GEMINI_API_KEY"
       },
       {
         "model": "gemini-2.5-flash",
-        "enabled": true,
+        "enabled": false,
         "rateLimited": false,
-        "usageCount": 9,
+        "usageCount": 10,
         "name": "gemini-2",
         "provider": "gemini",
         "key": "env:GEMINI_API_KEY_2"
       },
       {
         "model": "gemini-2.5-flash",
-        "enabled": true,
+        "enabled": false,
         "rateLimited": false,
-        "usageCount": 9,
+        "usageCount": 10,
         "name": "gemini-3",
         "provider": "gemini",
         "key": "env:GEMINI_API_KEY_3"
       },
       {
         "model": "gemini-2.5-flash",
-        "enabled": true,
+        "enabled": false,
         "rateLimited": false,
         "usageCount": 9,
         "name": "gemini-4",
         "provider": "gemini",
         "key": "env:GEMINI_API_KEY_4"
       },
+      {
+        "model": "nvidia/llama-3.3-nemotron-super-49b-v1",
+        "enabled": false,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "nvidia-1",
+        "provider": "nvidia",
+        "key": "env:NVIDIA_API_KEY"
+      },
+      {
+        "model": "meta/llama-3.3-70b-instruct",
+        "enabled": false,
+        "rateLimited": false,
+        "usageCount": 0,
+        "name": "nvidia-2",
+        "provider": "nvidia",
+        "key": "env:NVIDIA_API_KEY"
+      },
+      {
+        "model": "deepseek-ai/DeepSeek-V3.1",
+        "enabled": true,
+        "rateLimited": false,
+        "usageCount": 1061,
+        "name": "together-deepseek",
+        "provider": "custom",
+        "baseUrl": "https://api.together.xyz/v1/chat/completions",
+        "key": "env:TOGETHER_API_KEY"
+      },
       {
         "model": "meta-llama/llama-3.3-70b-instruct:free",
         "enabled": true,
         "rateLimited": false,
-        "usageCount": 0,
+        "usageCount": 77,
         "name": "openrouter-1",
         "provider": "openrouter",
         "key": "env:OPENROUTER_API_KEY"
@@ -106,7 +152,7 @@
         "model": "meta-llama/llama-3.3-70b-instruct:free",
         "enabled": true,
         "rateLimited": false,
-        "usageCount": 0,
+        "usageCount": 77,
         "name": "openrouter-2",
         "provider": "openrouter",
         "key": "env:OPENROUTER_API_KEY_2"
@@ -115,7 +161,7 @@
         "model": "meta-llama/llama-3.3-70b-instruct:free",
         "enabled": true,
         "rateLimited": false,
-        "usageCount": 0,
+        "usageCount": 78,
         "name": "openrouter-3",
         "provider": "openrouter",
         "key": "env:OPENROUTER_API_KEY_3"
@@ -124,21 +170,12 @@
         "model": "meta-llama/llama-3.1-405b-instruct",
         "enabled": true,
         "rateLimited": false,
-        "usageCount": 0,
+        "usageCount": 78,
         "name": "together-1",
         "provider": "custom",
         "baseUrl": "https://api.together.xyz/v1/chat/completions",
         "key": "env:TOGETHER_API_KEY"
       },
-      {
-        "model": "nvidia/llama-3.3-nemotron-super-49b-v1",
-        "enabled": true,
-        "rateLimited": false,
-        "usageCount": 0,
-        "name": "nvidia-1",
-        "provider": "nvidia",
-        "key": "env:NVIDIA_API_KEY"
-      },
       {
         "model": "gpt-4o-mini",
         "enabled": false,
@@ -191,6 +228,5 @@
   ],
   "cli": {
     "theme": "ember"
-  },
-  "primaryProvider": "groq-2"
+  }
 }

package/config/hardware.json

@@ -1,9 +1,9 @@
 {
   "gpu": "NVIDIA GeForce GTX 1060 6GB",
   "vramGB": 6,
-  "ramGB": 32,
+  "ramGB": 16,
   "platform": "windows",
   "cudaAvailable": true,
   "appleSilicon": false,
-  "detectedAt": "2026-04-29T18:49:39.763Z"
+  "detectedAt": "2026-05-01T19:50:20.289Z"
 }