open-agents-ai 0.184.5 → 0.184.7

package/README.md

@@ -26,16 +26,6 @@ An autonomous multi-turn tool-calling agent that reads your code, makes changes,
 
 An LLM is a high-bandwidth associative generative core — closer to a cortex-like prior than to a complete agent. Its weights contain broad latent structure, but they do not by themselves give you situated continuity, durable task state, calibrated action policies, or grounded memory management. Open Agents treats the model as one organ inside a larger organism. The framework provides the rest: sensors, effectors, memory stores, routing, gating, evaluation, and persistence.
 
-**Empirical proof** — 65 IQ-grade reasoning tasks (LSAT Logic Games, Bayesian probability, cryptarithmetic, constraint satisfaction, formal logic):
-
-| | With Framework (Full Organism) | Pure Reasoning (Cortex Only) |
-|---|---|---|
-| **4B model** | 65/65 (100%) | 15/15 (100%) |
-| **9B model** | 50/50 (100%) | 12/15 (80%) |
-| **30B model** | 20/20 (100%) | 11/15 (73%) |
-
-When the framework provides externalized computation (shell), persistent state (files), and verification loops (validators) — model size becomes irrelevant. A 4B model with the full agentic framework outperforms a 30B model reasoning alone.
-
 **What the framework provides:**
 
 | Layer | Biological Analog | Implementation |
@@ -109,6 +99,7 @@ The agent uses tools autonomously in a loop — reading errors, fixing code, and
 - **Task templates** — specialized system prompts and tool recommendations for code, document, analysis, plan tasks
 - **Inference capability scoring** — canirun.ai-style hardware assessment at first launch: memory/compute/speed scores, per-model compatibility matrix, recommended model selection
 - **Auto-install everything** — first-run wizard auto-installs Ollama, curl, Python3, python3-venv with platform-aware package managers (apt, dnf, yum, pacman, apk, zypper, brew)
+- **Sponsored inference** — `/sponsor` walks through a 5-step wizard to share your GPU with the world: select endpoints, choose banner animation (8 presets + AI-generated custom), set header message/links, configure transport (cloudflared/libp2p) + rate limits, and go live. Consumers discover sponsors via `/endpoint sponsor`. Secure proxy relay with per-IP rate limiting, daily token budgets, model allowlist, and concurrent request caps. Sponsor's raw API URL is never exposed. See [Sponsored Inference](#sponsored-inference--share-your-gpu-with-the-world) below
 - **P2P inference network** — `/expose` local models or forward any `/endpoint` (Chutes, Groq, OpenRouter, etc.) through the libp2p P2P mesh. Passthrough mode (`/expose passthrough`) relays upstream API requests; `--loadbalance` distributes rate-limited token budgets across peers. `/expose config` provides an arrow-key menu for all settings. Gateway stats show budget remaining from `x-ratelimit-*` headers. Background daemon persists across OA restarts
 - **P2P mesh networking** — `/p2p` with secret-safe variable placeholders (`{{OA_VAR_*}}`), trust tiers (LOCAL/TEE/VERIFIED/PUBLIC), WebSocket peer mesh, and inference routing with automatic secret redaction/injection
 - **Secret vault** — `/secrets` manages API keys and credentials with AES-256-GCM encrypted persistence; secrets are automatically redacted before sending to untrusted inference peers and re-injected on response
@@ -1446,6 +1437,70 @@ nexus(action='budget_set', auto_approve_below='0.01')  # Auto-approve micropayme
 - All outbound messages scanned for key material before sending
 - Keys NEVER appear in tool output, logs, or LLM context
 
+## Sponsored Inference — Share Your GPU With the World
+
+Anyone running Open Agents can become an inference sponsor — sharing their local models (or forwarded cloud endpoints) with users worldwide through a secure, branded relay.
+
+### For Sponsors: `/sponsor`
+
+Run `/sponsor` to walk through the 5-step onboarding wizard:
+
+```
+Step 1 → Select endpoints (auto-discovers local Ollama models + configured /endpoints)
+Step 2 → Choose banner animation (8 presets: wave, pulse, matrix, sparkle, radar, circuit, fire)
+         or generate a custom animation with your local LLM
+Step 3 → Set header message + clickable link (displayed to consumers during inference)
+Step 4 → Configure transport (cloudflared tunnel and/or libp2p P2P mesh)
+         + rate limits (req/min, tokens/day, max concurrent, model allowlist)
+Step 5 → Review and Go Live
+```
+
+**What happens under the hood:**
+- A secure reverse proxy starts on localhost, forwarding to your backend
+- Bearer token auth gate — unauthenticated requests rejected
+- Per-IP sliding window rate limiting + global daily token budget
+- Model allowlist enforcement (block models you don't want to share)
+- Token usage tracked from both Ollama and OpenAI response formats
+- Cloudflared tunnel creates a public HTTPS URL (or libp2p for decentralized relay)
+- Your raw API endpoint URL is **never exposed** — consumers only see the tunnel URL
+- Config persists to `.oa/sponsor/config.json` — survives restarts
+
+**Management:**
+```bash
+/sponsor          # Dashboard (when active) or wizard (when inactive)
+/sponsor status   # Usage metrics: requests, tokens, active connections, unique users
+/sponsor pause    # Stop serving, keep config
+/sponsor remove   # Retire sponsorship entirely
+```
+
+### For Consumers: `/endpoint sponsor`
+
+Users who need inference can discover and connect to sponsors:
+
+```bash
+/endpoint sponsor          # Browse available sponsored endpoints
+                           # Arrow-key select → auto-configures as active endpoint
+/endpoint <url> --auth <key>  # Direct connection with shared credentials
+```
+
+When using sponsored inference, the sponsor's banner animation and message appear in your header area.
+
+### Architecture
+
+```
+Consumer OA ──→ Cloudflared Tunnel ──→ Sponsor Proxy ──→ Ollama/vLLM
+                (HTTPS)                (auth + rate limit)   (local)
+                                       │
+                                       ├─ Bearer token gate
+                                       ├─ Per-IP sliding window (N req/min)
+                                       ├─ Daily token budget tracking
+                                       ├─ Model allowlist enforcement
+                                       ├─ Concurrent request cap
+                                       └─ Response header sanitization
+```
+
+The tunnel fix uses debounced restarts with exponential cooldown (10s → 20s → 40s), stopping auto-restart after 3 consecutive failures to prevent Cloudflare rate limiting. Progress indicators emit every 5 seconds during startup, and specific error messages are shown for common failure modes (ENOENT, port conflict, 429, DNS).
+
 ## Dream Mode — Creative Idle Exploration
 
 When you're not actively tasking the agent, Dream Mode lets it creatively explore your codebase and generate improvement proposals autonomously. The system models real human sleep architecture with four stages per cycle:

package/dist/index.js

@@ -42644,21 +42644,29 @@ async function stepBanner(config, rl, availableRows) {
   return false;
 }
 async function stepHeader(config, rl, availableRows) {
+  function msgDetail() {
+    return config.header.message || "(none set \u2014 press Enter to edit)";
+  }
+  function linkDetail() {
+    return config.header.linkUrl ? `${config.header.linkUrl}  [${config.header.linkText || "link"}]` : "(none set \u2014 press Enter to edit)";
+  }
   const items = [
     { key: "hdr", label: "Header Content" },
     {
       key: "message_toggle",
       label: `${config.header.messageEnabled ? "[x]" : "[ ]"} Header Message`,
-      detail: config.header.message || "(none set)"
+      detail: msgDetail()
     },
     {
       key: "link_toggle",
       label: `${config.header.linkEnabled ? "[x]" : "[ ]"} Link`,
-      detail: config.header.linkUrl ? `${config.header.linkUrl}  [${config.header.linkText || "link"}]` : "(none set)"
+      detail: linkDetail()
     },
     { key: "sep", label: "" },
     { key: "next", label: selectColors.green("  Next Step  \u2192") }
   ];
+  const msgIdx = 1;
+  const linkIdx = 2;
   const result = await tuiSelect({
     items,
     title: "Step 3/5 \u2014 Header Text & Links",
@@ -42679,31 +42687,52 @@ async function stepHeader(config, rl, availableRows) {
         }
       }
       return false;
+    },
+    // Use onCustomKey + helpers.getInput() for proper inline text capture
+    // (same pattern as TTS clone rename in voice list)
+    customKeyHint: "  e edit  space toggle",
+    onCustomKey: (item, key, helpers) => {
+      if ((key === "e" || key === "E") && item.key === "message_toggle") {
+        config.header.messageEnabled = true;
+        helpers.getInput("Header message:", config.header.message).then((text) => {
+          if (text !== null && text.trim()) {
+            config.header.message = text.trim().slice(0, 80);
+            helpers.updateItem(msgIdx, {
+              label: `[x] Header Message`,
+              detail: msgDetail()
+            });
+          }
+          helpers.render();
+        });
+        return true;
+      }
+      if ((key === "e" || key === "E") && item.key === "link_toggle") {
+        config.header.linkEnabled = true;
+        helpers.getInput("Link URL:", config.header.linkUrl).then((url) => {
+          if (url !== null && url.trim()) {
+            config.header.linkUrl = url.trim();
+            helpers.getInput("Link text:", config.header.linkText).then((text) => {
+              if (text !== null && text.trim()) {
+                config.header.linkText = text.trim();
+              }
+              helpers.updateItem(linkIdx, {
+                label: `[x] Link`,
+                detail: linkDetail()
+              });
+              helpers.render();
+            });
+          } else {
+            helpers.render();
+          }
+        });
+        return true;
+      }
+      return false;
     }
   });
   if (!result.confirmed)
     return false;
-  if (result.key === "next")
-    return true;
-  if (result.key === "message_toggle") {
-    config.header.messageEnabled = true;
-    const msg = await promptLine(rl, "Header message: ");
-    if (msg)
-      config.header.message = msg.slice(0, 80);
-    return stepHeader(config, rl, availableRows);
-  }
-  if (result.key === "link_toggle") {
-    config.header.linkEnabled = true;
-    const url = await promptLine(rl, "Link URL: ");
-    if (url) {
-      config.header.linkUrl = url;
-      const text = await promptLine(rl, "Link text: ");
-      if (text)
-        config.header.linkText = text;
-    }
-    return stepHeader(config, rl, availableRows);
-  }
-  return false;
+  return result.key === "next";
 }
 async function stepTransport(config, rl, availableRows) {
   const items = [
@@ -42734,12 +42763,16 @@ async function stepTransport(config, rl, availableRows) {
     { key: "sep", label: "" },
     { key: "next", label: selectColors.green("  Next Step  \u2192") }
   ];
+  const rpmIdx = items.findIndex((i) => i.key === "rpm");
+  const tpdIdx = items.findIndex((i) => i.key === "tpd");
+  const concIdx = items.findIndex((i) => i.key === "conc");
   const result = await tuiSelect({
     items,
     title: "Step 4/5 \u2014 Transport & Rate Limits",
     rl,
     skipKeys: ["hdr_transport", "hdr_limits", "sep"],
     availableRows,
+    customKeyHint: "  e edit  space toggle",
     onAction: (item, action) => {
       if (action === "space") {
         if (item.key === "cf_toggle") {
@@ -42754,6 +42787,42 @@ async function stepTransport(config, rl, availableRows) {
         }
       }
       return false;
+    },
+    onCustomKey: (item, key, helpers) => {
+      if ((key === "e" || key === "E") && item.key === "rpm") {
+        helpers.getInput("Max req/min:", String(config.rateLimits.maxRequestsPerMinute)).then((val) => {
+          if (val !== null) {
+            const n = parseInt(val, 10);
+            if (!isNaN(n) && n > 0)
+              config.rateLimits.maxRequestsPerMinute = n;
+          }
+          helpers.updateItem(rpmIdx, { label: `  Max requests/min:     ${config.rateLimits.maxRequestsPerMinute}` });
+        });
+        return true;
+      }
+      if ((key === "e" || key === "E") && item.key === "tpd") {
+        helpers.getInput("Max tokens/day:", String(config.rateLimits.maxTokensPerDay)).then((val) => {
+          if (val !== null) {
+            const n = parseInt(val, 10);
+            if (!isNaN(n) && n > 0)
+              config.rateLimits.maxTokensPerDay = n;
+          }
+          helpers.updateItem(tpdIdx, { label: `  Max tokens/day:       ${config.rateLimits.maxTokensPerDay.toLocaleString()}` });
+        });
+        return true;
+      }
+      if ((key === "e" || key === "E") && item.key === "conc") {
+        helpers.getInput("Max concurrent:", String(config.rateLimits.maxConcurrent)).then((val) => {
+          if (val !== null) {
+            const n = parseInt(val, 10);
+            if (!isNaN(n) && n > 0)
+              config.rateLimits.maxConcurrent = n;
+          }
+          helpers.updateItem(concIdx, { label: `  Max concurrent:       ${config.rateLimits.maxConcurrent}` });
+        });
+        return true;
+      }
+      return false;
     }
   });
   if (!result.confirmed)
@@ -42765,27 +42834,6 @@ async function stepTransport(config, rl, availableRows) {
     }
     return true;
   }
-  if (result.key === "rpm") {
-    const val = await promptLine(rl, "Max requests per minute: ");
-    const n = parseInt(val, 10);
-    if (!isNaN(n) && n > 0)
-      config.rateLimits.maxRequestsPerMinute = n;
-    return stepTransport(config, rl, availableRows);
-  }
-  if (result.key === "tpd") {
-    const val = await promptLine(rl, "Max tokens per day: ");
-    const n = parseInt(val, 10);
-    if (!isNaN(n) && n > 0)
-      config.rateLimits.maxTokensPerDay = n;
-    return stepTransport(config, rl, availableRows);
-  }
-  if (result.key === "conc") {
-    const val = await promptLine(rl, "Max concurrent requests: ");
-    const n = parseInt(val, 10);
-    if (!isNaN(n) && n > 0)
-      config.rateLimits.maxConcurrent = n;
-    return stepTransport(config, rl, availableRows);
-  }
   return false;
 }
 async function stepReview(config, rl, availableRows) {
@@ -42880,16 +42928,6 @@ async function runSponsorWizard(ctx) {
   renderInfo("Toggle: /sponsor status | Pause: /sponsor pause | Remove: /sponsor remove");
   return config;
 }
-function promptLine(rl, prompt) {
-  return new Promise((resolve35) => {
-    process.stdout.write(prompt);
-    const handler = (line) => {
-      rl.removeListener("line", handler);
-      resolve35(line.trim());
-    };
-    rl.on("line", handler);
-  });
-}
 var ANIM_PRESETS;
 var init_sponsor_wizard = __esm({
   "packages/cli/dist/tui/sponsor-wizard.js"() {
@@ -48972,17 +49010,37 @@ async function handleSponsoredEndpoint(ctx, local) {
 `);
   const sponsors = [];
   const gateway = ctx.getExposeGateway?.();
-  if (gateway && gateway.isActive && gateway.tunnelUrl) {
+  if (gateway && gateway.isActive) {
+    const gwUrl = gateway.tunnelUrl || `http://127.0.0.1:${gateway._proxyPort || 11434}`;
     sponsors.push({
       name: "Local Gateway",
-      url: gateway.tunnelUrl,
+      url: gwUrl,
       authKey: gateway.authKey,
       models: [],
-      // would need to probe
       limits: { rpm: 60, tpd: 1e5 },
       source: "local"
     });
   }
+  const { loadSponsorConfig: loadSpCfg } = await Promise.resolve().then(() => (init_sponsor_wizard(), sponsor_wizard_exports));
+  const spCfg = loadSpCfg(ctx.repoRoot ?? process.cwd());
+  if (spCfg && spCfg.status === "active") {
+    const epUrls = new Set(sponsors.map((s) => s.url));
+    for (const ep of spCfg.endpoints.filter((e) => e.enabled)) {
+      if (!epUrls.has(ep.url)) {
+        sponsors.push({
+          name: ep.label,
+          url: ep.url,
+          authKey: "",
+          models: ep.models,
+          limits: {
+            rpm: spCfg.rateLimits.maxRequestsPerMinute,
+            tpd: spCfg.rateLimits.maxTokensPerDay
+          },
+          source: "local"
+        });
+      }
+    }
+  }
   const sponsorDir2 = join58(ctx.repoRoot ?? process.cwd(), ".oa", "sponsor");
   const knownFile = join58(sponsorDir2, "known-sponsors.json");
   try {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "open-agents-ai",
-  "version": "0.184.5",
+  "version": "0.184.7",
   "description": "AI coding agent powered by open-source models (Ollama/vLLM) \u2014 interactive TUI with agentic tool-calling loop",
   "type": "module",
   "main": "./dist/index.js",