opencode-usage-coach 0.2.5 → 0.3.1

package/agents/usage-coach-harness.md

@@ -36,7 +36,12 @@ The user's message is the task source. If it has multiple distinct parts, decomp
 
 **Multi-model, 1 terminal:** the work runs via the plugin tools `generate` and `grade`, which use the configured models (generator/grader from harness.config.json) on the same server — so you get e.g. a paid generator + a free grader without a second terminal. Do NOT use the `task` tool for harness work; use `generate`/`grade`.
 
-**Parallel independent tasks:** if the decomposed tasks are INDEPENDENT, prefer `generate_batch` (one call, all results at once) over multiple sequential `generate` calls. Cap by quota coaching (big-OK → 3-4; throttle → 1-2; STOP → none). DEPENDENT tasks (B needs A) → sequential `generate` calls.
+**Quota-aware loop strategy (the tools handle this):** the quota coaching injected into your system prompt drives loop strategy. You do NOT pick the model — `generate`/`generate_batch` auto-switch to a lighter model on THROTTLE (if `lighterModel` is configured) and refuse on STOP. You pick the **task size and parallelism**:
+- **GO + headroom** → big tasks OK, use `generate_batch` for independent tasks (full parallel).
+- **THROTTLE** → small tasks only, sequential or limited parallelism. Tools cap concurrency at 2 and use a lighter model automatically.
+- **STOP** → finish the in-progress task, then halt. `generate_batch` returns an error on STOP — call `task_update(current, "halted_quota")` and stop.
+
+DEPENDENT tasks (B needs A) → always sequential `generate` calls, regardless of quota.
 
 1. Call `harness_start(name, N)` to register the run on the panel.
 2. For each task i (1..N):

package/dist/index.js

@@ -25,7 +25,7 @@ function setStateDir(dir) {
 }
 var NOOP_HOOKS = {};
 function log(msg) {
-  if (DEBUG) try {
+  try {
     appendFileSync(LOG_FILE, `${(/* @__PURE__ */ new Date()).toISOString()} ${msg}
 `);
   } catch {
@@ -80,50 +80,53 @@ async function runModel(client, model, prompt, directory) {
     const providerID = slash >= 0 ? model.slice(0, slash) : model;
     const modelID = slash >= 0 ? model.slice(slash + 1) : "";
     const s = await client.session.create({ body: { title: "uc-harness-sub" }, query: { directory } });
-    const id = s?.data?.info?.id;
-    if (!id) {
-      log(`runModel(${model}): no session id returned`);
-      return null;
-    }
+    log(`runModel(${model}): create resp = ${JSON.stringify(s?.data ?? s).slice(0, 500)}`);
+    const id = s?.data?.info?.id ?? s?.data?.id ?? s?.id;
+    if (!id) return `ERROR: session.create returned no id (response: ${JSON.stringify(s?.data ?? s).slice(0, 200)})`;
     log(`runModel(${model}): session ${id} created, prompt ${prompt.length} chars`);
     await client.session.prompt({
       path: { id },
       body: { model: { providerID, modelID }, parts: [{ type: "text", text: prompt }] }
     });
-    let lastStatus = "";
+    log(`runModel(${model}): prompt sent to ${id}`);
+    let lastMsgCount = -1;
     let timedOut = false;
+    let text = "";
     for (let i = 0; i < 600; i++) {
       await sleep(1e3);
-      const st = await client.session.status({ path: { id } });
-      const status = st?.data?.[id]?.status ?? st?.data?.status;
-      if (status !== lastStatus) {
-        log(`runModel(${model}): poll ${i}s status="${status}"`);
-        lastStatus = String(status);
-      }
-      if (status === "idle" || status === "completed") break;
-      if (status === "error" || status === "failed") {
-        log(`runModel(${model}): sub-session ended with "${status}"`);
-        break;
+      try {
+        const msgs = await client.session.messages({ path: { id } });
+        const all = msgs?.data ?? msgs ?? [];
+        if (all.length !== lastMsgCount) {
+          log(`runModel(${model}): poll[${i}] msgs=${all.length}`);
+          lastMsgCount = all.length;
+        }
+        const lastAssistant = all.filter((m) => (m?.info?.role ?? m?.role) === "assistant").pop();
+        const parts = lastAssistant?.parts ?? lastAssistant?.info?.parts ?? [];
+        const t = parts.filter((p) => p?.type === "text").map((p) => p?.text ?? "").join("").trim();
+        if (t) {
+          text = t;
+          log(`runModel(${model}): got assistant text at poll[${i}] (${t.length} chars)`);
+          break;
+        }
+      } catch (e) {
+        log(`runModel(${model}): poll[${i}] messages err: ${String(e).slice(0, 120)}`);
       }
       if (i === 599) timedOut = true;
     }
     const elapsed = Math.round((Date.now() - t0) / 1e3);
-    const msgs = await client.session.messages({ path: { id } });
-    const all = msgs?.data ?? [];
-    const lastAssistant = all.filter((m) => m?.info?.role === "assistant").pop();
-    const parts = lastAssistant?.parts ?? [];
-    const text = parts.filter((p) => p?.type === "text").map((p) => p?.text ?? "").join("");
     try {
       await client.session.remove?.({ path: { id } });
     } catch {
     }
     log(`runModel(${model}): done ${elapsed}s, ${text.length} chars${timedOut ? " [TIMEOUT]" : ""}`);
-    if (timedOut) return `[runModel TIMEOUT after ${elapsed}s \u2014 result may be incomplete]
-${text.trim()}`;
-    return text.trim() || null;
+    if (timedOut) return `[runModel TIMEOUT after ${elapsed}s \u2014 no assistant text appeared]
+${text}`;
+    return text || `ERROR: no assistant text after ${elapsed}s (${lastMsgCount} messages seen)`;
   } catch (e) {
-    log(`runModel err (${model}, ${Math.round((Date.now() - t0) / 1e3)}s): ${String(e)}`);
-    return null;
+    const elapsed = Math.round((Date.now() - t0) / 1e3);
+    log(`runModel err (${model}, ${elapsed}s): ${String(e)}`);
+    return `ERROR: runModel exception after ${elapsed}s: ${String(e)}`;
   }
 }
 var num = (e, d) => {
@@ -372,28 +375,53 @@ async function UsageCoachPlugin(input) {
             return "Harness complete.";
           }
         }),
-        // Per-role model execution (config-driven, same server, no deadlock).
+        // Per-role model execution (config-driven, quota-aware, same server, no deadlock).
+        // P1: quota decision drives model selection + concurrency.
         generate: tool({
-          description: "Run the GENERATOR model on a prompt. The generator can use tools (write/edit files). Returns the model's text response.",
+          description: "Run the GENERATOR model on a prompt. Quota-aware: on THROTTLE, auto-switches to lighterModel if configured. Returns the model's text response.",
           args: { prompt: tool.schema.string() },
           async execute(args, ctx) {
             const cfg = readHarnessCfg(ctx.directory);
             if (!cfg.generator) return 'ERROR: no generator model configured. Set "generator" in harness.config.json (see harness.config.example.json).';
-            const out = await runModel(input.client, cfg.generator, args.prompt, ctx.directory);
-            return out ?? `(generator ${cfg.generator} produced no text)`;
+            let decision = "GO";
+            try {
+              decision = current().decision;
+            } catch {
+            }
+            const throttle = decision === "THROTTLE" && cfg.lighterModel;
+            const model = throttle ? cfg.lighterModel : cfg.generator;
+            const out = await runModel(input.client, model, args.prompt, ctx.directory);
+            return out + (throttle ? `
+[usage-coach] quota THROTTLE \u2014 used lighter model ${cfg.lighterModel}` : "");
           }
         }),
         generate_batch: tool({
-          description: "Run the GENERATOR model on MULTIPLE tasks in PARALLEL. Pass an array of {id, prompt}. Returns all results at once. Use for INDEPENDENT tasks (much faster than sequential generate calls).",
+          description: "Run the GENERATOR model on MULTIPLE tasks. Quota-aware: GO = full parallel; THROTTLE = lighter model + concurrency capped at 2; STOP = refused. Use for INDEPENDENT tasks.",
           args: { tasks: tool.schema.array(tool.schema.object({ id: tool.schema.number(), prompt: tool.schema.string() })) },
           async execute(args, ctx) {
             const cfg = readHarnessCfg(ctx.directory);
             if (!cfg.generator) return 'ERROR: no generator model configured. Set "generator" in harness.config.json (see harness.config.example.json).';
-            const results = await Promise.all(args.tasks.map(async (t) => {
-              const out = await runModel(input.client, cfg.generator, t.prompt, ctx.directory);
-              return `[task ${t.id}] ${out ?? "(no output)"}`;
-            }));
-            return results.join("\n\n");
+            let decision = "GO";
+            try {
+              decision = current().decision;
+            } catch {
+            }
+            if (decision === "STOP") return 'ERROR: quota STOP \u2014 halt the harness loop now. Call task_update(current, "halted_quota") and stop.';
+            const throttle = decision === "THROTTLE" && cfg.lighterModel;
+            const model = throttle ? cfg.lighterModel : cfg.generator;
+            const limit = decision === "THROTTLE" ? 2 : args.tasks.length;
+            const results = [];
+            for (let i = 0; i < args.tasks.length; i += limit) {
+              const batch = args.tasks.slice(i, i + limit);
+              const out = await Promise.all(batch.map(async (t) => {
+                const r = await runModel(input.client, model, t.prompt, ctx.directory);
+                return `[task ${t.id}] ${r}`;
+              }));
+              results.push(...out);
+            }
+            const note = throttle ? `
+[usage-coach] quota THROTTLE \u2014 lighter model ${cfg.lighterModel}, concurrency capped at ${limit}` : "";
+            return results.join("\n\n") + note;
           }
         }),
         grade: tool({
@@ -404,7 +432,8 @@ async function UsageCoachPlugin(input) {
             const model = cfg.grader ?? cfg.generator;
             if (!model) return 'FAIL\n(ERROR: no grader/generator model configured. Set "grader" and "generator" in harness.config.json.)';
             const out = await runModel(input.client, model, args.prompt, ctx.directory);
-            if (!out) return "FAIL\n(grader produced no output \u2014 treating as FAIL to trigger a revise)";
+            if (out.startsWith("ERROR:")) return `FAIL
+(${out})`;
             const firstLine = out.split("\n").find((l) => l.trim()) ?? "";
             const f = firstLine.trim();
             const isPass = /^pass\b/i.test(f);

package/dist/tui.js

@@ -173,6 +173,7 @@ function initializeTui(api, disposeRoot) {
         const hf = join(STATE_DIR, sid, "harness.json");
         if (existsSync(hf)) h = JSON.parse(readFileSync(hf, "utf8"));
       }
+      if (!h || h.active === false) h = readHarness();
     } catch {
       h = null;
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-usage-coach",
-  "version": "0.2.5",
+  "version": "0.3.1",
   "description": "opencode closed-loop usage coach \u2014 quota SENSE -> coaching DECIDE -> loop ACT + TUI integration",
   "type": "module",
   "main": "./dist/index.js",