@inetafrica/open-claudia 3.1.0 → 3.1.2

package/CHANGELOG.md

@@ -1,5 +1,22 @@
 # Changelog
 
+## v3.1.2 — On-demand recall query + cold-walker fallback fix
+
+Two things fold in: a new `recall query` CLI so the working agent can ask long-term memory a question mid-task, and a latent reliability fix in the recall walker's fallback path.
+
+- **New `open-claudia recall query "<question>"`** — the agent's side door into the seed → graph → walker pipeline. Where per-turn recall seeds on the *user's* message, this seeds the graph with the agent's OWN working context (via `--context`), walks the typed-edge graph with the **active-provider** judge, and returns the kept pack sections with why-bullets and pointers. It closes the documented cold-start hole: mid-build, when the vocabulary you're using shares no words with the pack that governs the work, ordinary recall never resurfaces it — an explicit query carrying your working context does. Read-only (no edge reinforcement), no pre-gate (an explicit ask always runs full), and fail-open (a failed judge returns the whole candidate list unjudged rather than nothing). Flags: `--context`, `--json`, `--no-walker`, `--no-episodes`, `--limit`, `--timeout`.
+- **Cold-walker outputSchema 400 fix** — the recall walker prefers a warm persistent judge (Claude only) and falls back to a cold spawn otherwise. The cold path was passing an array-typed `outputSchema`, which the Anthropic API turns into a custom-tool `input_schema` and rejects with `400 input_schema.type: Input should be 'object'`. So any time the warm walker errored and fell back, recall silently 400'd and failed-open — judgement quietly disabled on the fallback path. Removed the schema entirely (the warm path never sent one, and `walk()` already extracts the JSON array from freeform text, filters to known ids, and clamps scores), so freeform is the proven contract on both paths.
+- **Agent-space / OpenClaw:** new `core/recall/query.js` (`query()` + `mergeMatches`), `excerptFor` exported from the discoverer, `recall query` wired into `bin/recall.js`. New hermetic `test-recall-query.js` in the pretest chain (unjudged fail-open, judged keep-only, judged-empty, garbage fail-open, context-seed merge, blank-question reject).
+
+## v3.1.1 — /logout and switch-account: re-auth is finally reachable from chat
+
+Reported live: with Claude or Codex already connected, there was **no way to sign into a different account** — `/login` showed "connected" and tapping a model just switched models, because the auth-aware picker (correctly, since v3.0.36) only starts a login flow when a provider is *not* authenticated. And `/clear_oauth_token` alone didn't help: it clears the bot-held token but the CLI credential store keeps `authStatus` reading "authenticated", so the picker still never offered a sign-in. Re-auth needs a *full* sign-out first — this release adds it.
+
+- **New `/logout [claude|codex]`** (owner-only) — a full sign-out: runs `claude auth logout` / `codex logout` against the CLI credential store **and** (for Claude) clears the bot-held OAuth token from `.env`, process env, and vault in the same step. Partial clears were exactly the trap. Reports each step honestly (including "CLI not found — store untouched"), then offers a **Sign in** button so the fresh login is one tap away. Without an argument it shows provider buttons.
+- **`/login` picker grows a 🔁 Switch account button** under each *connected* provider. Tapping it asks first — showing who is currently signed in (email where the CLI exposes it, else the auth method: OAuth token source / ChatGPT account / API key) and warning that the bot loses that provider until the new login completes — then signs out and chains straight into the fresh login flow. A stray tap can't kill a working auth: the destructive step is behind the explicit confirm.
+- **`/clear_oauth_token` is now documented as the narrow tool** (bot token only); `/logout claude` is the full story.
+- **Agent-space / OpenClaw:** new auth-flow primitives `logoutClaude`/`logoutCodex`/`providerLogout`/`canLogout`/`providerAccountSummary` (core/auth-flow.js), shared executor `performProviderLogout` + `startProviderLogin` (core/handlers.js), new callback family `sw:`/`lo:`/`losw:`/`li:` (core/actions.js, owner-gated). New test `test-logout.js` in the pretest chain.
+
 ## v3.1.0 — Memory becomes required: safe-mode boot, crash-loop rescue, and /downgrade
 
 Three waves folded into one release, graduated to a minor because boot semantics change. Wave 0: the memory substrate is now **required** — a missing `node:sqlite` boots into safe mode instead of silently degrading, a boot crash loop lands there too, and a new `/downgrade` completes the rescue-from-chat story. Wave 1: the memory-substrate revival, root-caused from a live incident — a pod ran 16 days repeating corrected branding mistakes because its entire memory substrate silently never initialized. Wave 2: the first fix batch from the 2026-07-20 blind code review (18 verified findings; deeper architectural items ship in later batches).

package/README.md

@@ -208,10 +208,11 @@ When you select a project, the last conversation is automatically resumed. Tap "
 | Command | Description |
 |---------|-------------|
 | `/auth_status` | Reports redacted Claude auth status and whether an OAuth token is configured |
-| `/login` | Starts a Claude login flow, sends the URL/code, accepts paste-back codes (`/auth_code <code>`, `/cancel_auth`) |
+| `/login` | Starts a Claude login flow, sends the URL/code, accepts paste-back codes (`/auth_code <code>`, `/cancel_auth`). Connected providers show a **🔁 Switch account** button that signs out and starts a fresh login |
+| `/logout [claude\|codex]` | Full sign-out: clears the CLI credential store **and** (for Claude) the bot-held OAuth token, then offers a sign-in button. Without an argument, shows provider buttons |
 | `/setup_token` | Runs `claude setup-token`; the resulting OAuth token is stored without echoing it |
 | `/use_oauth_token [<token>]` | Stores `CLAUDE_CODE_OAUTH_TOKEN` for non-interactive runs; the message is deleted when possible. Without an argument, send the token as the next message |
-| `/clear_oauth_token` | Removes the stored OAuth token |
+| `/clear_oauth_token` | Removes the stored OAuth token only (`/logout claude` for a full sign-out) |
 
 Tokens are redacted from chat output and logs.
 

package/bin/recall.js

@@ -1,16 +1,26 @@
 // open-claudia recall-stats [--json]
 // open-claudia recall graph [--sync] [--json]
+// open-claudia recall query "<question>" [--context "<text>"] [--json]
 //
 // Inspect the recall system: discoverer metrics (precision/miss-rescue/noise,
-// latency, cost) and the typed-edge graph (node/edge counts, sync).
+// latency, cost), the typed-edge graph (node/edge counts, sync), and the
+// on-demand walker query (ask long-term memory a question mid-task).
 
 const path = require("path");
 
 const HELP = `
-Inspect the memory recall system.
+Inspect and query the memory recall system.
 
   open-claudia recall-stats [--json]      Discoverer metrics summary
   open-claudia recall graph [--sync]      Recall-graph stats (optionally re-sync structural edges first)
+  open-claudia recall query "<question>"  On-demand memory lookup: FTS seeds + graph expansion +
+                                          walker judgement → kept sections with why-bullets
+    --context "<text>"   Working context to seed alongside the question (what you're doing now)
+    --json               Machine-readable result
+    --no-walker          Skip the LLM judge; print the raw seed+graph candidate list
+    --no-episodes        Skip past-conversation transcript hits
+    --limit <n>          Pack seed width (default 6)
+    --timeout <ms>       Walker timeout (default 25000)
 
 The discoverer engine logs one line per turn to recall-metrics.jsonl and a
 rolling summary to recall-metrics-summary.json. Switch engines per chat with
@@ -59,10 +69,66 @@ function graph(args) {
   for (const [type, n] of Object.entries(s.byType || {})) console.log(`  ${type}: ${n}`);
 }
 
+async function queryCmd(args) {
+  const flags = { context: "", json: false, noWalker: false, noEpisodes: false, limit: 0, timeout: 0 };
+  const positional = [];
+  for (let i = 0; i < args.length; i++) {
+    const a = args[i];
+    if (a === "--json") flags.json = true;
+    else if (a === "--no-walker") flags.noWalker = true;
+    else if (a === "--no-episodes") flags.noEpisodes = true;
+    else if (a === "--context") flags.context = args[++i] || "";
+    else if (a === "--limit") flags.limit = Number(args[++i]);
+    else if (a === "--timeout") flags.timeout = Number(args[++i]);
+    else positional.push(a);
+  }
+  const question = positional.join(" ").trim();
+  if (!question) {
+    console.error('Usage: open-claudia recall query "<question>" [--context "<text>"] [--json] [--no-walker] [--no-episodes] [--limit <n>] [--timeout <ms>]');
+    process.exit(1);
+  }
+  // The walker reads its timeout from env at module load — set before require.
+  if (flags.timeout > 0) process.env.RECALL_DISCOVERER_TIMEOUT_MS = String(flags.timeout);
+  const lib = require(path.join(__dirname, "..", "core", "recall", "query"));
+  let res;
+  try {
+    res = await lib.query(question, {
+      context: flags.context,
+      walker: !flags.noWalker,
+      episodes: !flags.noEpisodes,
+      limit: flags.limit || undefined,
+    });
+  } catch (e) {
+    console.error(`recall query failed: ${e.message}`);
+    process.exit(1);
+  }
+  if (flags.json) { console.log(JSON.stringify(res, null, 2)); return; }
+  const w = res.walker || {};
+  console.log(`Recall query — ${res.results.length} kept of ${res.candidates} candidate(s) (seeds: ${res.seeds.packs} pack, ${res.seeds.entities} entity, ${res.seeds.tools} tool; +${res.activated} graph-activated)`);
+  console.log(res.judged
+    ? `Walker: ${w.provider || "?"}${w.model ? `/${w.model}` : ""} (${w.source}, ${w.ms}ms)`
+    : `Walker: ${w.source === "off" ? "off" : "failed"} — UNJUDGED candidate list (fail-open)`);
+  for (const r of res.results) {
+    console.log("");
+    const score = r.score == null ? "" : ` (score ${r.score})`;
+    const viaTag = r.activated ? ` [via ${r.via || "graph"}]` : "";
+    console.log(`■ ${r.kind}: ${r.name}${score}${viaTag}`);
+    if (r.why) console.log(`  why: ${r.why}`);
+    if (r.kind === "pack" && r.lines && r.lines.length) {
+      for (const l of r.lines) console.log(`  > [${l.section}:${l.lineno}] ${l.text}`);
+    } else if (r.excerpt) {
+      console.log(`  > ${r.excerpt.replace(/\n/g, " ").slice(0, 300)}`);
+    }
+    if (r.pointer) console.log(`  ↳ ${r.pointer}`);
+  }
+  if (!res.results.length) console.log('\n(nothing relevant — rephrase, add --context, or use --no-walker to see raw candidates)');
+}
+
 function run(args) {
   const sub = args[0];
   if (sub === "--help" || sub === "help") { console.log(HELP); process.exit(0); }
   if (sub === "graph") { graph(args.slice(1)); return; }
+  if (sub === "query") { queryCmd(args.slice(1)); return; }
   // default / "stats": metrics summary
   stats(args.filter((a) => a !== "stats"));
 }

package/core/actions.js

@@ -22,6 +22,7 @@ const { finishOnboarding } = require("./onboarding");
 const {
   startSession, switchProviderState, startNewConversation,
   endCurrentSession, activeContinueOptions, applyProviderSetting,
+  performProviderLogout, startProviderLogin,
 } = require("./handlers");
 const providerRegistry = require("./providers");
 const jobs = require("./jobs");
@@ -29,6 +30,7 @@ const scheduler = require("./scheduler");
 const { sideChatCoordinator } = require("./side-chat");
 const {
   claudeProviderAuthStatus, runClaudeAuthCommand, runCodexLoginStatus, runCodexDeviceLogin,
+  canLogout, providerAccountSummary,
 } = require("./auth-flow");
 
 const REMOVED_PROVIDER_CALLBACK_MESSAGE = "This button is from an older Open Claudia release. Cursor Agent support has been removed. Open /backend to choose Claude Code or OpenAI Codex.";
@@ -350,6 +352,26 @@ async function handleAction(envelope) {
     await send(switched ? `Switched to ${beLabel}.\nModel: ${model}` : `Model: ${model}`);
     return;
   }
+  // Switch-account / sign-out / sign-in buttons (from the /login picker and
+  // /logout). sw: asks first — a stray tap must not kill a working auth.
+  if (d.startsWith("sw:") || d.startsWith("lo:") || d.startsWith("losw:") || d.startsWith("li:")) {
+    if (!isChatOwner(envelope.channelId)) return send("Owner only — provider auth is shared across users.");
+    const providerId = d.slice(d.indexOf(":") + 1);
+    if (!canLogout(providerId)) return send(`No sign-out flow for "${providerId}". Open /login again.`);
+    const label = providerId === "claude" ? "Claude" : "Codex";
+    if (d.startsWith("sw:")) {
+      const account = providerAccountSummary(providerId);
+      return send(
+        `${label} is signed in as: ${account}\n\nSign out and start a fresh sign-in? The bot loses ${label} until the new login completes.`,
+        { keyboard: { inline_keyboard: [[
+          { text: `Yes — switch ${label} account`, callback_data: `losw:${providerId}` },
+          { text: "Cancel", callback_data: "noop" },
+        ]] } },
+      );
+    }
+    if (d.startsWith("li:")) return startProviderLogin(providerId);
+    return performProviderLogout(providerId, { thenLogin: d.startsWith("losw:") });
+  }
   if (d.startsWith("m:")) {
     state.settings.model = d.slice(2) === "default" ? null : d.slice(2);
     await send(`Model: ${state.settings.model || "default"}`);

package/core/auth-flow.js

@@ -494,6 +494,69 @@ async function runCodexDeviceLogin() {
   });
 }
 
+// ── Logout / switch account ────────────────────────────────────────
+// Full sign-out clears BOTH credential sources. A partial clear is the
+// re-auth trap: authStatus stays "authenticated" via the other source, so
+// the /login picker never offers a fresh sign-in (the v3.0.36 side effect).
+
+function logoutClaude(deps = {}) {
+  const steps = [];
+  const tokenInfo = (deps.getClaudeOAuthToken || getClaudeOAuthToken)();
+  if (tokenInfo.value) {
+    (deps.clearClaudeOAuthToken || clearClaudeOAuthToken)();
+    steps.push(`Bot OAuth token cleared (${tokenInfo.source})`);
+  }
+  const binary = deps.binary !== undefined ? deps.binary : claudeExecutable();
+  if (!binary) {
+    steps.push("Claude CLI not found — CLI credential store untouched");
+    return { ok: steps.length > 1 || !!tokenInfo.value, steps };
+  }
+  const result = (deps.runCommand || runCommand)(binary, ["auth", "logout"], { env: claudeSubprocessEnv(), timeout: 15000 });
+  steps.push(result.ok ? "Claude CLI store signed out" : `Claude CLI logout: ${summarizeAuthOutput(result.output, false)}`);
+  return { ok: result.ok || !!tokenInfo.value, steps };
+}
+
+function logoutCodex(deps = {}) {
+  const binary = deps.binary !== undefined ? deps.binary : codexExecutable();
+  if (!binary) return { ok: false, steps: ["Codex CLI not found — nothing to sign out"] };
+  const result = (deps.runCommand || runCommand)(binary, ["logout"], { env: codexSubprocessEnv(), timeout: 15000 });
+  return { ok: result.ok, steps: [result.ok ? "Codex CLI store signed out" : `Codex logout: ${summarizeAuthOutput(result.output, false)}`] };
+}
+
+const PROVIDER_LOGOUTS = { claude: logoutClaude, codex: logoutCodex };
+
+function canLogout(providerId) {
+  return Object.prototype.hasOwnProperty.call(PROVIDER_LOGOUTS, providerId);
+}
+
+function providerLogout(providerId, deps = {}) {
+  const fn = PROVIDER_LOGOUTS[providerId];
+  if (!fn) return { ok: false, steps: [`No sign-out flow for provider "${providerId}"`] };
+  return fn(deps);
+}
+
+// Short "who is signed in" line for confirm prompts — best effort, only
+// what the CLI exposes.
+function providerAccountSummary(providerId) {
+  if (providerId === "claude") {
+    const tokenInfo = getClaudeOAuthToken();
+    if (tokenInfo.value) return `OAuth token (${tokenInfo.source})`;
+    const output = runClaudeAuthStatusDiagnostic();
+    const email = (String(output).match(/[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}/i) || [null])[0];
+    return email || "CLI credential store";
+  }
+  if (providerId === "codex") {
+    const status = runCodexLoginStatus();
+    if (!status.ok) return "not signed in";
+    const email = (String(status.output).match(/[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,}/i) || [null])[0];
+    if (email) return email;
+    if (/chatgpt/i.test(status.output)) return "ChatGPT account";
+    if (/api key/i.test(status.output)) return "API key";
+    return "CLI credential store";
+  }
+  return "unknown";
+}
+
 module.exports = {
   CLAUDE_OAUTH_TOKEN_KEY, CLAUDE_OAUTH_VAULT_KEY,
   getClaudeOAuthToken,
@@ -527,4 +590,9 @@ module.exports = {
   sendCodexAuthStatusSummary,
   saveCodexApiKeyWithCli,
   runCodexDeviceLogin,
+  logoutClaude,
+  logoutCodex,
+  canLogout,
+  providerLogout,
+  providerAccountSummary,
 };

package/core/handlers.js

@@ -48,6 +48,7 @@ const {
   runClaudeAuthCommand, clearPendingClaudeAuth,
   sendCodexAuthStatusSummary, runCodexDeviceLogin, clearPendingCodexAuth,
   saveCodexApiKeyWithCli,
+  canLogout, providerLogout, providerAccountSummary,
 } = require("./auth-flow");
 const providerRegistry = require("./providers");
 
@@ -1086,13 +1087,16 @@ function buildModelPicker({ authAware = false, state = currentState(), registry
   const rows = [];
   for (const provider of registry.listProviders()) {
     let authMarker = "";
+    let connected = false;
     if (authAware) {
-      let connected = false;
       try { connected = provider.authStatus().state === "authenticated"; } catch (_) {}
       authMarker = connected ? " · connected" : " · tap to connect";
     }
     const availableMarker = providerAvailability(provider) ? "" : " · unavailable";
     rows.push([{ text: `── ${provider.label}${availableMarker}${authMarker} ──`, callback_data: "noop" }]);
+    if (authAware && connected && canLogout(provider.id)) {
+      rows.push([{ text: "🔁 Switch account", callback_data: `sw:${provider.id}` }]);
+    }
     const choices = provider.modelChoices();
     for (let index = 0; index < choices.length; index += 3) {
       rows.push(choices.slice(index, index + 3).map((model) => ({
@@ -1928,6 +1932,43 @@ register({
   },
 });
 
+async function startProviderLogin(providerId) {
+  if (providerId === "claude") return runClaudeAuthCommand(["auth", "login", "--claudeai"], "Claude login");
+  if (providerId === "codex") return runCodexDeviceLogin();
+  return send(`No login flow for provider "${providerId}".`);
+}
+
+async function performProviderLogout(providerId, { thenLogin = false } = {}) {
+  const provider = providerFromRegistry(providerId);
+  const label = provider ? provider.label : providerId;
+  const result = providerLogout(providerId);
+  await send(`${label} sign-out:\n${result.steps.map((s) => `• ${s}`).join("\n")}`);
+  if (thenLogin) return startProviderLogin(providerId);
+  await send(`Sign in with a different account when ready — tap below or use /login.`, {
+    keyboard: { inline_keyboard: [[{ text: `Sign in ${label}`, callback_data: `li:${providerId}` }]] },
+  });
+}
+
+module.exports.startProviderLogin = startProviderLogin;
+module.exports.performProviderLogout = performProviderLogout;
+
+register({
+  name: "logout", description: "Sign out of a provider (full: CLI store + bot token)", args: "[claude|codex]", ownerOnly: true,
+  handler: async (env, { tail }) => {
+    if (!ownerEnv(env)) return send("Owner only — provider auth is shared across users.");
+    const raw = (tail || "").trim().toLowerCase();
+    const target = ({ gpt: "codex", openai: "codex", chatgpt: "codex", anthropic: "claude" })[raw] || raw;
+    if (canLogout(target)) return performProviderLogout(target);
+    if (target) return send(`Unknown provider "${target}". Use /logout claude or /logout codex.`);
+    return send("Sign out of which provider? This clears the CLI credential store (and Claude's bot token) so /login offers a fresh sign-in.", {
+      keyboard: { inline_keyboard: [[
+        { text: "Claude", callback_data: "lo:claude" },
+        { text: "Codex", callback_data: "lo:codex" },
+      ]] },
+    });
+  },
+});
+
 register({
   name: "setup_token", description: "Create Claude OAuth token", ownerOnly: true,
   handler: async (env) => {

package/core/recall/discoverer.js

@@ -282,6 +282,13 @@ async function runWalker(prompt, options = {}) {
       };
     } catch (e) { /* fall back to cold spawn below */ }
   }
+  // No outputSchema: a provider structured-output schema becomes a custom-tool
+  // input_schema, which the Anthropic API requires to be an OBJECT — the
+  // walker's top-level array schema is rejected 400 ("input_schema.type: Input
+  // should be 'object'"), so the cold path used to fail-open on every fallback.
+  // The warm path never sends a schema and the walk() parser already extracts
+  // the JSON array from freeform text, filters to known ids, and clamps scores,
+  // so freeform is the proven contract on both paths.
   const result = await runUtility(prompt, {
     purpose: "recall",
     provider: selection.providerId,
@@ -291,14 +298,6 @@ async function runWalker(prompt, options = {}) {
     legacyClaudeModelEnv: "RECALL_DISCOVERER_MODEL",
     systemPrompt: WALKER_SYSTEM,
     readOnly: true,
-    outputSchema: {
-      type: "array",
-      items: {
-        type: "object",
-        required: ["id"],
-        properties: { id: { type: "string" }, why: { type: "string" }, score: { type: "number" } },
-      },
-    },
     timeoutMs: WALKER_TIMEOUT_MS,
     cwd: options.cwd || process.cwd(),
   });
@@ -779,4 +778,4 @@ async function run(ctx) {
   };
 }
 
-module.exports = { name: "discoverer", run, runWalker, needsRecall, recallTier, walk, cascadeVerdict, cokeptPairs, scoutToolGaps, renderToolGaps, episodesAllowedForSpeaker };
+module.exports = { name: "discoverer", run, runWalker, needsRecall, recallTier, walk, excerptFor, cascadeVerdict, cokeptPairs, scoutToolGaps, renderToolGaps, episodesAllowedForSpeaker };

package/core/recall/query.js

@@ -0,0 +1,242 @@
+"use strict";
+
+// On-demand recall: the working agent's side door into the same
+// seed → graph-expansion → walker pipeline the discoverer runs per turn.
+// `open-claudia recall query "<question>" [--context "<working context>"]`
+//
+// The per-turn engine seeds from the USER's message; mid-build the agent's
+// actual question often shares zero words with the governing pack (measured
+// 2026-07-16: the embeddings-measurement stance never resurfaced during the
+// build it governed). Here the agent seeds with its own question + working
+// context, and graph edges + walker judgement bridge the vocabulary gap.
+//
+// Deliberate differences from the per-turn engine:
+//   - no pre-gate: an explicit query is always intentional (always "full").
+//   - read-only: never reinforces edges, never warms co-kept pairs — a
+//     query is a lookup, not a turn.
+//   - cold walker only: a one-shot CLI process must not own a warm child.
+//   - fail-open returns ALL candidates marked unjudged — the caller is an
+//     agent that can judge for itself; an empty answer helps nobody.
+
+const graph = require("./graph");
+const metrics = require("./metrics");
+const tuning = require("./tuning");
+const discoverer = require("./discoverer");
+
+let transcriptIndex = null;
+try { transcriptIndex = require("../transcript-index"); } catch (e) {}
+
+const COLD_ONLY_WARM = { isEnabled: () => false };
+const EXCERPT_CHARS = 600;
+
+function mergeMatches(primary, secondary, keyOf) {
+  const out = primary.map((m) => ({ ...m, origin: "user" }));
+  const seen = new Set(primary.map(keyOf));
+  for (const m of secondary) {
+    if (seen.has(keyOf(m))) continue;
+    out.push({ ...m, origin: "context" });
+  }
+  return out;
+}
+
+function clip(s, n) { return s && s.length > n ? s.slice(0, n) + "…" : (s || ""); }
+
+async function query(question, opts = {}) {
+  const packsLib = opts.packsLib || require("../packs");
+  const entitiesLib = opts.entitiesLib || require("../entities");
+  const toolsLib = opts.toolsLib || require("../tools");
+  const started = Date.now();
+  const q = String(question || "").trim();
+  if (!q) throw new Error("recall query needs a question");
+  const context = String(opts.context || "").trim();
+  const packLimit = Number(opts.limit) > 0 ? Math.min(Number(opts.limit), 12) : 6;
+
+  // Seed via FTS on the question, then the working context (deduped).
+  let packSeeds = [];
+  let entSeeds = [];
+  try {
+    packSeeds = mergeMatches(
+      packsLib.matchPacks(q, { limit: packLimit }),
+      context ? packsLib.matchPacks(context, { limit: packLimit }) : [],
+      (m) => m.dir,
+    );
+  } catch (e) {}
+  try {
+    entSeeds = mergeMatches(
+      entitiesLib.matchEntities(q, { limit: 4 }),
+      context ? entitiesLib.matchEntities(context, { limit: 4 }) : [],
+      (m) => m.slug,
+    );
+  } catch (e) {}
+  const toolSeeds = new Map();
+  try {
+    for (const t of toolsLib.matchTools(q, { limit: 5 })) toolSeeds.set(t.name, { ...t, origin: "user" });
+    if (context) {
+      for (const t of toolsLib.matchTools(context, { limit: 5 })) {
+        if (!toolSeeds.has(t.name)) toolSeeds.set(t.name, { ...t, origin: "context" });
+      }
+    }
+  } catch (e) {}
+
+  // Spreading activation from seeds across the typed-edge graph.
+  const seedNodes = [
+    ...packSeeds.map((m) => ({ id: `pack:${m.dir}`, score: m.score || 2 })),
+    ...entSeeds.map((m) => ({ id: `entity:${m.slug}`, score: m.score || 2 })),
+  ];
+  const seedIds = new Set(seedNodes.map((s) => s.id));
+  let activated = new Map();
+  if (graph.available()) {
+    try { graph.syncFromCorpus(packsLib, entitiesLib); } catch (e) {}
+    try { activated = graph.expand(seedNodes, {}); } catch (e) {}
+  }
+
+  // Candidates: question seeds by score, context seeds, then graph-activated.
+  const seedRank = new Map();
+  for (const m of packSeeds) seedRank.set(`pack:${m.dir}`, { score: m.score || 2, origin: m.origin });
+  for (const m of entSeeds) seedRank.set(`entity:${m.slug}`, { score: m.score || 2, origin: m.origin });
+  const orderedIds = [...seedIds].sort((a, b) => {
+    const ra = seedRank.get(a) || {}, rb = seedRank.get(b) || {};
+    const ua = ra.origin === "user" ? 0 : 1, ub = rb.origin === "user" ? 0 : 1;
+    return ua - ub || (rb.score || 0) - (ra.score || 0);
+  });
+  for (const [id] of [...activated.entries()].sort((x, y) => (y[1].activation || 0) - (x[1].activation || 0))) {
+    if (!seedIds.has(id)) orderedIds.push(id);
+  }
+  const candidates = [];
+  for (const id of orderedIds.slice(0, tuning.get("walkerMaxCandidates"))) {
+    const base = discoverer.excerptFor(id, packsLib, entitiesLib);
+    if (!base) continue;
+    const act = activated.get(id);
+    candidates.push({ ...base, activated: !seedIds.has(id), via: act ? act.via : null });
+  }
+
+  // Tools: direct seeds plus tools declared by any candidate pack.
+  const candPackDirs = new Set();
+  for (const c of candidates) if (c.id.startsWith("pack:")) candPackDirs.add(c.id.slice(5));
+  const toolCand = new Map();
+  for (const [name] of toolSeeds) toolCand.set(name, { name, via: null });
+  if (candPackDirs.size) {
+    try {
+      for (const t of toolsLib.listTools()) {
+        if (t.pack && candPackDirs.has(t.pack) && !toolCand.has(t.name)) {
+          toolCand.set(t.name, { name: t.name, via: `pack:${t.pack}` });
+        }
+      }
+    } catch (e) {}
+  }
+  for (const c of toolCand.values()) {
+    const base = discoverer.excerptFor(`tool:${c.name}`, packsLib, entitiesLib);
+    if (!base) continue;
+    candidates.push({ ...base, activated: !!c.via, via: c.via });
+  }
+
+  // Episodes: FTS over past transcripts. The CLI has no speaker context, so
+  // the R6 owner gate is open by design (the CLI runs as the owner on their
+  // own box); bot-side callers still pass through the per-turn engine's gate.
+  if (opts.episodes !== false && transcriptIndex && transcriptIndex.available()) {
+    try {
+      const { hits } = transcriptIndex.search(q, { limit: tuning.get("episodeLimit") });
+      for (const h of (hits || [])) {
+        const terms = [...String(h.snip || "").matchAll(/>>(.+?)<</g)].map((m) => m[1]).slice(0, 4);
+        candidates.push({
+          id: `episode:${require("path").basename(String(h.file || ""), ".jsonl")}:${h.line}`,
+          name: `Past conversation${h.project ? ` (${h.project})` : ""}${h.ts ? ` ${String(h.ts).slice(0, 10)}` : ""}`,
+          description: "",
+          excerpt: clip(String(h.snip || "").replace(/>>|<</g, ""), EXCERPT_CHARS),
+          activated: true,
+          via: "transcript",
+          _terms: terms,
+        });
+      }
+    } catch (e) { /* episodic recall is best-effort */ }
+  }
+
+  // Walker judgement (cold spawn only). opts.walker === false skips the spawn.
+  let walkRes = { kept: null, scores: null, source: "off", walkerMs: 0, provider: null, model: null, costUsd: null };
+  if (opts.walker !== false && candidates.length) {
+    walkRes = await discoverer.walk(q, context, candidates, {
+      warmWalker: COLD_ONLY_WARM,
+      cwd: opts.cwd,
+      ...(opts.walkerOptions || {}),
+    });
+  }
+  const whyMap = walkRes.kept;
+  const scores = walkRes.scores;
+
+  const results = [];
+  for (const c of candidates) {
+    if (whyMap && !whyMap.has(c.id)) continue;
+    const kind = c.id.startsWith("pack:") ? "pack"
+      : c.id.startsWith("entity:") ? "entity"
+      : c.id.startsWith("tool:") ? "tool" : "episode";
+    const r = {
+      id: c.id,
+      kind,
+      name: c.name,
+      why: whyMap ? whyMap.get(c.id) || "" : "",
+      score: scores && scores.has(c.id) ? scores.get(c.id) : null,
+      activated: !!c.activated,
+      via: c.via || null,
+      excerpt: c.excerpt || "",
+    };
+    if (kind === "pack") {
+      const dir = c.id.slice(5);
+      r.pointer = `open-claudia pack show ${dir}`;
+      try {
+        const hits = packsLib.searchLines([q, context].filter(Boolean).join(" "), { dir, k: 8 });
+        if (hits && hits.length) r.lines = hits.map((h) => ({ section: h.section, lineno: h.lineno, text: h.text }));
+      } catch (e) { /* a pack with no line hits still shows its excerpt */ }
+    } else if (kind === "entity") {
+      r.pointer = `open-claudia entity show ${c.id.slice(7)}`;
+    } else if (kind === "tool") {
+      r.pointer = `open-claudia tool show ${c.id.slice(5)}`;
+    } else {
+      const qterms = (c._terms && c._terms.length
+        ? c._terms.join(" ")
+        : String(c.excerpt).split(/\s+/).slice(0, 4).join(" ")).replace(/"/g, "");
+      r.pointer = `open-claudia transcript-search "${qterms}" --all`;
+    }
+    results.push(r);
+  }
+  results.sort((a, b) => (b.score ?? -1) - (a.score ?? -1));
+
+  const out = {
+    question: q,
+    hasContext: !!context,
+    seeds: { packs: packSeeds.length, entities: entSeeds.length, tools: toolSeeds.size },
+    activated: activated.size,
+    candidates: candidates.length,
+    judged: !!whyMap,
+    walker: {
+      source: walkRes.source,
+      ms: walkRes.walkerMs || 0,
+      provider: walkRes.provider || null,
+      model: walkRes.model || null,
+    },
+    results,
+    latencyMs: Date.now() - started,
+  };
+
+  try {
+    metrics.logTurn({
+      engine: "query",
+      provider: walkRes.provider || "",
+      model: walkRes.model || "",
+      query: q,
+      tier: "query",
+      seeds: seedNodes,
+      activated: [...activated.entries()].map(([id, a]) => ({ id, activation: a.activation, hop: a.hop })),
+      cand: candidates.map((c) => ({ id: c.id, act: !!c.activated })),
+      kept: results.map((r) => ({ id: r.id, why: r.why, score: r.score })),
+      gated: false,
+      latencyMs: out.latencyMs,
+      walkerMs: walkRes.walkerMs,
+      walker: walkRes.source,
+      costUsd: walkRes.costUsd || 0,
+    });
+  } catch (e) {}
+
+  return out;
+}
+
+module.exports = { query, mergeMatches };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inetafrica/open-claudia",
-  "version": "3.1.0",
+  "version": "3.1.2",
   "description": "An always-on, provider-agnostic coding-agent harness for Claude Code and OpenAI Codex via chat",
   "main": "bot.js",
   "bin": {
@@ -9,8 +9,8 @@
   "scripts": {
     "setup": "node setup.js",
     "start": "node bot.js",
-    "pretest": "node test-provider-fixture.js && node test-provider-core-prompt.js && node test-turn-observer.js && node test-provider-prompt-parity.js && node test-provider-env-isolation.js && node test-provider-resume-parity.js && node test-provider-bootstrap.js && node test-provider-registry.js && node test-provider-stream-decoder.js && node test-provider-events.js && node test-provider-claude.js && node test-provider-codex.js && node test-provider-runner.js && node test-provider-refusal-recovery.js && node test-provider-run-lifecycle.js && node test-provider-gateway.js && node test-provider-side-chat.js && node test-single-runtime.js && node test-provider-migration-backup.js && node test-provider-state-migration.js && node test-provider-session-history.js && node test-provider-session-commands.js && node test-provider-compaction.js && node test-provider-scheduler.js && node test-credential-capture.js && node test-cursor-removal.js && node test-utility-provider-policy.js && node test-provider-subagent.js && node test-provider-recall.js && node test-provider-pack-review.js && node test-memory-mutation-queue.js && node test-provider-dream.js && node test-provider-enforcer.js && node test-provider-tool-hooks.js && node test-provider-boot-matrix.js && node test-provider-capabilities.js && node test-provider-language.js && node test-provider-coupling-audit.js && node test-provider-parity-e2e.js && node test-kazee-channel-health.js && node test-prompt-budget.js && node test-agency-ledger.js && node test-agency-deliberate.js && node test-guardrail.js && node test-spaces-comment-content.js && node test-boot-health.js && node test-lexical-floor.js && node test-safe-mode.js",
-    "test": "OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node -e \"require('./vault'); console.log('OK')\" && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-runner-watchdog-static.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-usage-accounting.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-engine.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-graph.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-discoverer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-walker-score.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-score-gate.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-cokept-warming.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-project-transcripts-smoke.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-abilities.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-extraction.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-couse.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-transfer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-tiers.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-merge-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-learning-e2e.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-pack-nesting.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-pack-search.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-extractive-injection.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-journal-dedupe.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-read-signal.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tools.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-graph.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tooling-mode.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-manifest.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-evolution.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-approval-async.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-unified-identity.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-queue-routing.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-persona-packs.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-speaker-persona.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-persona-pipeline.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-relationship-gate.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-relationship.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-enforcer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-identity-prune.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-kazee-message-id.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-delivery-contract.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-run-lock.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-web-sessions.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-poll-recovery.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-409-grace.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-conflict-heal-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-single-instance.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-streaming-split.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-spaces-autoenable.js"
+    "pretest": "node test-provider-fixture.js && node test-provider-core-prompt.js && node test-turn-observer.js && node test-provider-prompt-parity.js && node test-provider-env-isolation.js && node test-provider-resume-parity.js && node test-provider-bootstrap.js && node test-provider-registry.js && node test-provider-stream-decoder.js && node test-provider-events.js && node test-provider-claude.js && node test-provider-codex.js && node test-provider-runner.js && node test-provider-refusal-recovery.js && node test-provider-run-lifecycle.js && node test-provider-gateway.js && node test-provider-side-chat.js && node test-single-runtime.js && node test-provider-migration-backup.js && node test-provider-state-migration.js && node test-provider-session-history.js && node test-provider-session-commands.js && node test-provider-compaction.js && node test-provider-scheduler.js && node test-credential-capture.js && node test-cursor-removal.js && node test-utility-provider-policy.js && node test-provider-subagent.js && node test-provider-recall.js && node test-provider-pack-review.js && node test-memory-mutation-queue.js && node test-provider-dream.js && node test-provider-enforcer.js && node test-provider-tool-hooks.js && node test-provider-boot-matrix.js && node test-provider-capabilities.js && node test-provider-language.js && node test-provider-coupling-audit.js && node test-provider-parity-e2e.js && node test-kazee-channel-health.js && node test-prompt-budget.js && node test-agency-ledger.js && node test-agency-deliberate.js && node test-guardrail.js && node test-spaces-comment-content.js && node test-boot-health.js && node test-lexical-floor.js && node test-safe-mode.js && node test-logout.js",
+    "test": "OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node -e \"require('./vault'); console.log('OK')\" && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-runner-watchdog-static.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-usage-accounting.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-engine.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-graph.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-discoverer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-query.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-walker-score.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-score-gate.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-cokept-warming.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-project-transcripts-smoke.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-abilities.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-extraction.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-couse.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-transfer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-tiers.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-ability-merge-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-learning-e2e.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-pack-nesting.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-pack-search.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-extractive-injection.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-journal-dedupe.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-read-signal.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tools.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-graph.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tooling-mode.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-tool-manifest.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-evolution.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-approval-async.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-unified-identity.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-queue-routing.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-persona-packs.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-speaker-persona.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-persona-pipeline.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-recall-relationship-gate.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-relationship.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-enforcer.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-identity-prune.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-kazee-message-id.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-delivery-contract.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-run-lock.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-web-sessions.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-poll-recovery.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-409-grace.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-telegram-conflict-heal-guard.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-single-instance.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-streaming-split.js && OPEN_CLAUDIA_TEST=1 WORKSPACE=/tmp/open-claudia-test CLAUDE_PATH=./test-fixtures/fake-agent-cli.js FAKE_AGENT_KIND=claude AGENT_ENV_PASSTHROUGH=FAKE_AGENT_KIND node test-spaces-autoenable.js"
   },
   "files": [
     "bot.js",

package/test-provider-language.js

@@ -146,7 +146,28 @@ assert.match(pkg.description, /provider-agnostic coding-agent harness/i);
 // New /downgrade rolls back to the previously-running version from boot
 // history (strict semver-validated target); /upgrade//downgrade clear the
 // crash counter as deliberate fix attempts. package.json engines >=24.
-assert.strictEqual(pkg.version, "3.1.0", "release version must remain unchanged without explicit approval");
+// 3.1.1 approved by Sumeet 2026-07-20: he reported the re-auth trap ("wjen i
+// want to reauth a new account for say claude or gpt and i do /login and
+// select it but i am still logged in it just shows me the same model instead
+// of a reatuh and there is no logut option?") and green-lit the proposed fix
+// (deferred once behind the smart-brenda slash-menu check, then resumed).
+// Ships: /logout [claude|codex] full sign-out — CLI credential store AND the
+// bot-held Claude token in one step (partial clears were the trap: any
+// surviving source keeps authStatus "authenticated" so the /login picker
+// never re-offers a sign-in); a 🔁 Switch account button on connected
+// providers in /login behind an explicit confirm that names the signed-in
+// account where the CLI exposes it, then chains logout → fresh login.
+// 3.1.2 approved by Sumeet 2026-07-20 ("If everything you have works and is
+// ready fold it in and push all out in a new version"): on-demand recall query
+// CLI (`open-claudia recall query "<question>"`) — seeds the graph with the
+// agent's OWN working context (not the user's words), walks it with the
+// active-provider judge, returns pack sections + why-bullets; closes the
+// cold-start hole where mid-build vocabulary shared no words with the governing
+// pack. Also fixes a latent cold-walker bug: the warm→cold fallback sent an
+// array-typed outputSchema the Anthropic API rejects 400, so every fallback
+// silently failed-open — removed; freeform JSON parse is the proven contract on
+// both paths.
+assert.strictEqual(pkg.version, "3.1.2", "release version must remain unchanged without explicit approval");
 for (const keyword of ["claude", "codex", "coding-agent", "provider-agnostic"]) {
   assert.ok(pkg.keywords.includes(keyword), `package keyword missing: ${keyword}`);
 }