opencode-usage-coach 0.3.2 → 0.3.4

package/README.md

@@ -166,7 +166,7 @@ Recurring issues and fixes — mostly learned the hard way during development.
 - **Export must be `{ tui }`** (an object) — a bare function is misread as a server plugin.
 - **Color prop is `fg`** (not `foreground`): `style={{ fg: theme.current.success }}`.
 - **Call `codexbar` via `spawn`** — the `$` BunShell leaks command output into the TUI.
-- **Harness not visible?** The TUI scans session subdirectories for the most recent `active:true` harness. A finished harness (`active:false`) is hidden.
+- **Harness not visible?** The TUI shows the **current session's** harness only. A finished harness (`active:false`) is hidden. A harness started in another session won't appear here — switch to that session to see it.
 
 ### LLM model selection (generate/grade)
 - **`generator` is required in `harness.config.json`** — the tools return a clear error if missing (the old z.ai fallback was removed in v0.2.4).
@@ -187,13 +187,19 @@ Recurring issues and fixes — mostly learned the hard way during development.
 - `(no output)` means the sub-session returned no text. Check `~/.cache/opencode-usage-coach/coach.log` for the runModel trace (requires `UC_DEBUG=1`).
 - **Note**: runModel only terminates on explicit `idle`/`completed` status. An earlier bug broke on `!status` (undefined) immediately — fixed.
 
+### generate / generate_batch returns "Tool execution aborted"
+- **Cause:** opencode imposes a tool execution timeout (~60–120s, not configurable). `runModel` polls the sub-session for up to 10 min; if the generator model takes longer than the tool timeout, opencode aborts the call.
+- **This is a platform limit, not a plugin bug** — there is no config key to extend it.
+- **Mitigation:** split large tasks into smaller ones that finish within the timeout. The NEXT directives + revise loop help — a FAIL triggers a focused revise rather than a monolithic retry.
+- The sub-session keeps running in the background after abort; its file writes are preserved. Only the tool's return value is lost.
+
 ### Multi-session (per-session state isolation)
 - Each opencode session has a unique sessionID; harness state is isolated per-session:
   ```
   ~/.cache/opencode-usage-coach/projects/<dir-hash>/<sessionID>/harness.json
   ```
 - Different working directories get separate project state (keyed by path hash).
-- The TUI auto-discovers the most recent `active:true` harness — switching sessions shows that session's harness.
+- The TUI shows the **current session's** harness only (per-session isolation) — no cross-session leakage. A harness running in session B does not appear in session A's panel.
 - Harness completion sets `active:false` → hidden from the TUI.
 - Override the state path with `UC_STATE_DIR` (forces global state).
 

package/dist/index.js

@@ -38,6 +38,21 @@ function writeState(c) {
   } catch {
   }
 }
+function rulesFile() {
+  return join(STATE_DIR, "rules.md");
+}
+function failuresFile() {
+  return join(STATE_DIR, "failures.ndjson");
+}
+function readRules() {
+  try {
+    const f = rulesFile();
+    if (!existsSync(f)) return "";
+    return readFileSync(f, "utf8").trim();
+  } catch {
+    return "";
+  }
+}
 function harnessFile(sessionID) {
   return join(STATE_DIR, sessionID || "_default", "harness.json");
 }
@@ -73,56 +88,28 @@ function readHarnessCfg(dir) {
   };
 }
 async function runModel(client, model, prompt, directory) {
-  const sleep = (ms) => new Promise((r) => setTimeout(r, ms));
   const t0 = Date.now();
   try {
     const slash = model.indexOf("/");
     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 } });
-    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 ${prompt.length} chars`);
-    await client.session.prompt({
+    log(`runModel(${model}): session ${id} created, sending prompt (${prompt.length} chars)`);
+    const resp = await client.session.prompt({
       path: { id },
       body: { model: { providerID, modelID }, parts: [{ type: "text", text: prompt }] }
     });
-    log(`runModel(${model}): prompt sent to ${id}`);
-    let lastMsgCount = -1;
-    let timedOut = false;
-    let text = "";
-    for (let i = 0; i < 600; i++) {
-      await sleep(1e3);
-      try {
-        const msgs = await client.session.messages({ path: { id } });
-        const all = msgs?.data ?? msgs ?? [];
-        if (all.length !== lastMsgCount) {
-          log(`runModel(${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: ${String(e).slice(0, 120)}`);
-      }
-      if (i === 599) timedOut = true;
-    }
     const elapsed = Math.round((Date.now() - t0) / 1e3);
+    const parts = resp?.data?.parts ?? resp?.parts ?? [];
+    const text = parts.filter((p) => p?.type === "text").map((p) => p?.text ?? "").join("");
     try {
       await client.session.remove?.({ path: { id } });
     } catch {
     }
-    log(`runModel(${model}): done ${elapsed}s, ${text.length} chars${timedOut ? " [TIMEOUT]" : ""}`);
-    if (timedOut) return `[runModel TIMEOUT after ${elapsed}s \u2014 no assistant text appeared]
-${text}`;
-    return text || `ERROR: no assistant text after ${elapsed}s (${lastMsgCount} messages seen)`;
+    log(`runModel(${model}): done ${elapsed}s, ${text.length} chars`);
+    return text.trim() || `ERROR: no assistant text in prompt response after ${elapsed}s (parts: ${parts.length}, types: ${parts.map((p) => p?.type).join(",")})`;
   } catch (e) {
     const elapsed = Math.round((Date.now() - t0) / 1e3);
     log(`runModel err (${model}, ${elapsed}s): ${String(e)}`);
@@ -376,7 +363,7 @@ Then: harness_done(). Follow the [usage-coach NEXT] directive each tool returns.
           async execute(args, ctx) {
             const h = readHarness(ctx.sessionID) ?? { name: "batch", total: 0, current: 0, tasks: [], usage: {}, active: true };
             h.tasks = h.tasks.filter((x) => x.id !== args.id);
-            h.tasks.push({ id: args.id, title: args.title, status: args.status, model: args.model ?? "", revisions: args.revisions ?? 0, score: args.score ?? null });
+            h.tasks.push({ id: args.id, title: args.title, status: args.status, model: args.model ?? "", revisions: args.revisions ?? 0, score: args.score ?? null, startedAt: (/* @__PURE__ */ new Date()).toISOString() });
             if (args.id > h.current) h.current = args.id;
             writeHarness(ctx.sessionID, h);
             return `task ${args.id} -> ${args.status}${args.score ? ` (${args.score})` : ""}`;
@@ -395,6 +382,110 @@ Then: harness_done(). Follow the [usage-coach NEXT] directive each tool returns.
             return "Harness complete.";
           }
         }),
+        record_failure: tool({
+          description: "Stage 1 (RECORD) of the learning loop. Append a failure record to failures.ndjson for later root-cause analysis.",
+          args: {
+            task: tool.schema.string(),
+            prompt: tool.schema.string(),
+            gradeResult: tool.schema.string(),
+            model: tool.schema.string().optional(),
+            revisions: tool.schema.number().optional()
+          },
+          async execute(args, _ctx) {
+            const rec = { ts: (/* @__PURE__ */ new Date()).toISOString(), task: args.task, prompt: args.prompt, gradeResult: args.gradeResult, model: args.model, revisions: args.revisions };
+            try {
+              mkdirSync(STATE_DIR, { recursive: true });
+              appendFileSync(failuresFile(), JSON.stringify(rec) + "\n");
+            } catch (e) {
+              log(`record_failure err: ${String(e)}`);
+            }
+            return `Failure recorded. [usage-coach NEXT] call investigate({failure: ${JSON.stringify(rec)}}) to find the root cause.`;
+          }
+        }),
+        investigate: tool({
+          description: "Stage 2 (INVESTIGATE) of the learning loop. Run the generator to analyze the ROOT CAUSE of a failure (not just the symptom).",
+          args: {
+            task: tool.schema.string(),
+            prompt: tool.schema.string(),
+            gradeResult: tool.schema.string(),
+            model: tool.schema.string().optional()
+          },
+          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 rcaPrompt = `A task failed. Analyze the ROOT CAUSE (not just the symptom).
+Task: ${args.task}
+What was expected (from grade): ${args.gradeResult}
+Read relevant files in the directory if needed.
+Output a structured root cause:
+category: (one of: constraint-violation, missing-context, tool-misuse, model-limitation, other)
+explanation: <why it failed>
+evidence: <file/line or specific quote>`;
+            const out = await runModel(input.client, cfg.generator, rcaPrompt, ctx.directory);
+            return out + "\n[usage-coach NEXT] call verify_diagnosis with this diagnosis.";
+          }
+        }),
+        verify_diagnosis: tool({
+          description: "Stage 3 (VERIFY) of the learning loop. Run the grader to check whether a diagnosis is CORRECT and ACTIONABLE (leads to a useful rule). Returns PASS/FAIL + a [usage-coach NEXT] directive.",
+          args: {
+            diagnosis: tool.schema.string(),
+            task: tool.schema.string(),
+            gradeResult: tool.schema.string()
+          },
+          async execute(args, ctx) {
+            const cfg = readHarnessCfg(ctx.directory);
+            const model = cfg.grader ?? cfg.generator;
+            if (!model) return "FAIL\n(ERROR: no grader/generator model configured.)\n[usage-coach NEXT] configure grader in harness.config.json, then retry verify_diagnosis.";
+            const verifyPrompt = `Verify this root-cause analysis for a failure.
+Task: ${args.task}
+Grade feedback: ${args.gradeResult}
+Diagnosis: ${args.diagnosis}
+Is the diagnosis CORRECT and ACTIONABLE (leads to a useful rule)?
+Output PASS (the diagnosis is right) or FAIL (re-investigate needed), then reason.`;
+            const out = await runModel(input.client, model, verifyPrompt, ctx.directory);
+            let verdict = "FAIL";
+            if (!out.startsWith("ERROR:")) {
+              const f = (out.split("\n").find((l) => l.trim()) ?? "").trim();
+              if (/^pass\b/i.test(f)) verdict = "PASS";
+              else verdict = "FAIL";
+            }
+            const next = verdict === "PASS" ? `
+[usage-coach NEXT] call generalize with this verified diagnosis.` : `
+[usage-coach NEXT] FAIL \u2014 re-investigate the root cause.`;
+            return out + "\n" + next;
+          }
+        }),
+        generalize: tool({
+          description: "Stage 4 (GENERALIZE) of the learning loop. Run the generator to turn a verified root cause into a reusable rule and append it to rules.md, so the next generate call includes it. Returns the rule text and a [usage-coach NEXT] directive.",
+          args: {
+            diagnosis: tool.schema.string(),
+            task: 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 genPrompt = `Turn this verified root cause into a GENERAL, REUSABLE rule for future tasks of this kind.
+Diagnosis: ${args.diagnosis}
+Failed task: ${args.task}
+Output a single rule in the form: 'For <task-type> tasks, always <check/do X> because <reason>.'
+Keep it concrete and actionable.`;
+            const out = await runModel(input.client, cfg.generator, genPrompt, ctx.directory);
+            const rule = out;
+            try {
+              const date = (/* @__PURE__ */ new Date()).toISOString().slice(0, 10);
+              mkdirSync(STATE_DIR, { recursive: true });
+              appendFileSync(rulesFile(), `## Rule (${date})
+${rule}
+Origin: ${args.task}
+
+`);
+            } catch (e) {
+              log(`generalize err: ${String(e)}`);
+            }
+            return `${rule}
+[usage-coach NEXT] rule saved to rules.md. The next generate call will include it. Call task_update for the original failed task -> failed, then proceed.`;
+          }
+        }),
         // Per-role model execution (config-driven, quota-aware, same server, no deadlock).
         // P1: quota decision drives model selection + concurrency.
         generate: tool({
@@ -410,7 +501,14 @@ Then: harness_done(). Follow the [usage-coach NEXT] directive each tool returns.
             }
             const throttle = decision === "THROTTLE" && cfg.lighterModel;
             const model = throttle ? cfg.lighterModel : cfg.generator;
-            const out = await runModel(input.client, model, args.prompt, ctx.directory);
+            const rules = readRules();
+            const prefix = rules ? `Lessons learned from previous failures (apply where relevant):
+${rules}
+
+---
+
+` : "";
+            const out = await runModel(input.client, model, prefix + args.prompt, ctx.directory);
             return out + (throttle ? `
 [usage-coach] quota THROTTLE \u2014 used lighter model ${cfg.lighterModel}` : "") + `
 [usage-coach NEXT] call task_update(i, title, "grading"), then grade to evaluate this work.`;
@@ -462,7 +560,12 @@ Then: harness_done(). Follow the [usage-coach NEXT] directive each tool returns.
             }
             const next = verdict === "PASS" ? `
 [usage-coach NEXT] PASS -> call task_update(i, title, "completed", "PASS"), then proceed to next task (or harness_done if last).` : `
-[usage-coach NEXT] FAIL -> if revisions < 2: task_update(i, title, "revising", revisions+1) + generate({prompt: "Apply feedback:\\n{grade result}\\nTask: {title}"}); else: task_update(i, title, "failed", "FAIL") -> next task.`;
+[usage-coach NEXT] FAIL -> if revisions < 2: task_update(i, title, "revising", revisions+1) + generate({prompt: "Apply feedback:\\n{grade result}\\nTask: {title}"}); else: run the learning loop before failing \u2014
+  1. record_failure({task, prompt, gradeResult, model, revisions})
+  2. investigate({task, prompt, gradeResult}) -> diagnosis
+  3. verify_diagnosis({diagnosis, task, gradeResult}) -> if PASS: generalize({diagnosis, task}) (saves rule to rules.md)
+  4. task_update(i, title, "failed", "FAIL") -> next task.
+The next generate call will automatically include the new rule.`;
             return out + "\n" + next;
           }
         })

package/dist/tui.js

@@ -5,7 +5,7 @@ import { effect as _$effect } from "@opentui/solid";
 import { createTextNode as _$createTextNode } from "@opentui/solid";
 import { insertNode as _$insertNode } from "@opentui/solid";
 import { createElement as _$createElement } from "@opentui/solid";
-import { readFileSync, existsSync, writeFileSync, mkdirSync, readdirSync, statSync } from "fs";
+import { readFileSync, existsSync, writeFileSync, mkdirSync, readdirSync, statSync, appendFileSync } from "fs";
 import { createHash } from "crypto";
 import { homedir } from "os";
 import { join, resolve } from "path";
@@ -91,21 +91,43 @@ function barFill(p) {
 function barEmpty(p) {
   return "\u2591".repeat(10 - Math.max(0, Math.min(10, Math.round(p / 10))));
 }
-function short(s, n) {
-  return s.length <= n ? s : s.slice(0, n - 1) + "\u2026";
-}
 function initializeTui(api, disposeRoot) {
   STATE_DIR = process.env.UC_STATE_DIR ?? projectStateDir(api.state.path.directory);
   STATE_FILE = join(STATE_DIR, "state.json");
   HARNESS_FILE = join(STATE_DIR, "harness.json");
   MARKER = join(STATE_DIR, "tui-loaded.txt");
+  const TUI_LOG = join(STATE_DIR, "tui-debug.log");
+  const tlog = (msg) => {
+    try {
+      appendFileSync(MARKER, `${(/* @__PURE__ */ new Date()).toISOString()} ${msg}
+`);
+      appendFileSync(TUI_LOG, `${(/* @__PURE__ */ new Date()).toISOString()} ${msg}
+`);
+    } catch (e) {
+      try {
+        appendFileSync(MARKER, `TLOG ERR: ${String(e)}
+`);
+      } catch {
+      }
+    }
+  };
   try {
     mkdirSync(STATE_DIR, {
       recursive: true
     });
-    writeFileSync(MARKER, `loaded ${(/* @__PURE__ */ new Date()).toISOString()} @ ${api.state.path.directory}`);
+    writeFileSync(MARKER, `loaded-v2 ${(/* @__PURE__ */ new Date()).toISOString()} @ ${api.state.path.directory}`);
   } catch {
   }
+  tlog(`init start | dir=${api.state.path.directory} | STATE_DIR=${STATE_DIR}`);
+  try {
+    tlog(`api keys=${Object.keys(api).join(",")}`);
+    tlog(`api.state=${JSON.stringify(api.state).slice(0, 400)}`);
+    tlog(`api.state.path=${JSON.stringify(api.state?.path).slice(0, 300)}`);
+    const r = api.route;
+    tlog(`api.route type=${typeof r} keys=${r && typeof r === "object" ? Object.keys(r).join(",") : "?"} val=${JSON.stringify(r).slice(0, 400)}`);
+  } catch (e) {
+    tlog(`api probe err: ${String(e)}`);
+  }
   const [getState, setState] = createSignal(readState());
   const [getHarness, setHarness] = createSignal(readHarness());
   const timer = setInterval(() => {
@@ -168,12 +190,14 @@ function initializeTui(api, disposeRoot) {
     }
     let h = null;
     try {
-      const sid = ctx.session_id ?? "";
+      const routeSid = api.route?.current?.params?.sessionID ?? "";
+      const sid = routeSid || (ctx.session_id ?? "");
       if (sid) {
         const hf = join(STATE_DIR, sid, "harness.json");
         if (existsSync(hf)) h = JSON.parse(readFileSync(hf, "utf8"));
+      } else {
+        h = readHarness();
       }
-      if (!h || h.active === false) h = readHarness();
     } catch {
       h = null;
     }
@@ -358,6 +382,8 @@ function initializeTui(api, disposeRoot) {
         const lbl = TLABEL[t.status] ?? t.status;
         const rev = t.revisions > 0 && t.status === "revising" ? `(${t.revisions})` : "";
         const mdl = t.model ? ` ${t.model.split("/").pop() ?? t.model}` : "";
+        const elapsed = t.startedAt ? Math.max(0, Math.round((Date.now() - new Date(t.startedAt).getTime()) / 1e3)) : 0;
+        const elapsedStr = t.status === "completed" || t.status === "failed" ? "" : elapsed > 0 ? ` ${elapsed}s` : "";
         nodes.push((() => {
           var _el$59 = _$createElement("text"), _el$60 = _$createTextNode(` \u25CF `), _el$61 = _$createTextNode(` `), _el$62 = _$createTextNode(` `);
           _$insertNode(_el$59, _el$60);
@@ -367,15 +393,18 @@ function initializeTui(api, disposeRoot) {
           _$insert(_el$59, mdl, _el$61);
           _$insert(_el$59, lbl, _el$62);
           _$insert(_el$59, rev, _el$62);
-          _$insert(_el$59, () => short(t.title, 12), null);
+          _$insert(_el$59, elapsedStr, _el$62);
+          _$insert(_el$59, () => t.title, null);
           _$effect((_$p) => _$setProp(_el$59, "style", st(sKey), _$p));
           return _el$59;
         })());
         const pv = t.model ? (t.model.split("/")[0] ?? "").split("-")[0] : "";
-        const q = pv && h.quotas?.[pv] ? h.quotas[pv] : null;
-        const pct = q ? q.fiveHour : 0;
+        const provCoach = s?.providers?.find((p) => p.id === pv || pv && p.id.startsWith(pv) || pv && pv.startsWith(p.id));
+        const rawPct = provCoach ? provCoach.fiveHour : s?.fiveHour ?? 0;
+        const pct = rawPct < 0 ? 0 : rawPct;
+        const pctLabel = rawPct < 0 ? "n/a" : `${rawPct}%`;
         nodes.push((() => {
-          var _el$63 = _$createElement("box"), _el$64 = _$createElement("text"), _el$66 = _$createElement("text"), _el$67 = _$createElement("text"), _el$68 = _$createElement("text"), _el$69 = _$createTextNode(` `), _el$70 = _$createTextNode(`%`);
+          var _el$63 = _$createElement("box"), _el$64 = _$createElement("text"), _el$66 = _$createElement("text"), _el$67 = _$createElement("text"), _el$68 = _$createElement("text"), _el$69 = _$createTextNode(` `);
           _$insertNode(_el$63, _el$64);
           _$insertNode(_el$63, _el$66);
           _$insertNode(_el$63, _el$67);
@@ -385,8 +414,7 @@ function initializeTui(api, disposeRoot) {
           _$insert(_el$66, () => barFill(pct));
           _$insert(_el$67, () => barEmpty(pct));
           _$insertNode(_el$68, _el$69);
-          _$insertNode(_el$68, _el$70);
-          _$insert(_el$68, pct, _el$70);
+          _$insert(_el$68, pctLabel, null);
           _$effect((_p$) => {
             var _v$1 = st("text"), _v$10 = st("textMuted");
             _v$1 !== _p$.e && (_p$.e = _$setProp(_el$66, "style", _v$1, _p$.e));
@@ -401,28 +429,32 @@ function initializeTui(api, disposeRoot) {
       }
     }
     return (() => {
-      var _el$71 = _$createElement("box");
-      _$setProp(_el$71, "flexDirection", "column");
-      _$insert(_el$71, nodes);
-      return _el$71;
+      var _el$70 = _$createElement("box");
+      _$setProp(_el$70, "flexDirection", "column");
+      _$insert(_el$70, nodes);
+      return _el$70;
     })();
   };
+  tlog("registering slots");
   api.slots.register({
     order: 80,
     slots: {
       sidebar_footer(ctx) {
+        tlog("sidebar_footer slot called");
         try {
           return panel(ctx);
-        } catch {
+        } catch (e) {
+          tlog(`sidebar_footer err: ${String(e)}`);
           return (() => {
-            var _el$72 = _$createElement("text");
-            _$insertNode(_el$72, _$createTextNode(`usage-coach`));
-            return _el$72;
+            var _el$71 = _$createElement("text");
+            _$insertNode(_el$71, _$createTextNode(`usage-coach`));
+            return _el$71;
           })();
         }
       }
     }
   });
+  tlog("slots registered, init complete");
 }
 var tui = async (api) => {
   createRoot((disposeRoot) => initializeTui(api, disposeRoot));

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-usage-coach",
-  "version": "0.3.2",
+  "version": "0.3.4",
   "description": "opencode closed-loop usage coach \u2014 quota SENSE -> coaching DECIDE -> loop ACT + TUI integration",
   "type": "module",
   "main": "./dist/index.js",