@hireobi/cli 0.1.8 → 0.1.9

package/README.md

@@ -1,24 +1,24 @@
 # Obi CLI
 
-Obi is an AI Search Employee. This package installs the `obi` CLI and the Obi Skill, which runs inside the assistant you already use: Claude Code, Cursor or Codex. The Skill does the work in the chat: public business research saved through a versioned work boundary, buyer questions, and server-measured engine answers kept apart from your assistant's own reasoning. Create your account at [agent.hireobi.com](https://agent.hireobi.com); learn more at [hireobi.com](https://hireobi.com).
+Obi is an AI Search Employee. This package installs the `obi` CLI and the Obi Skill, which runs inside the assistant you already use: Claude Code, Cursor, Codex or OpenCode. The Skill does the work in the chat: public business research saved through a versioned work boundary, buyer questions, and server-measured engine answers kept apart from your assistant's own reasoning. Create your account at [agent.hireobi.com](https://agent.hireobi.com); learn more at [hireobi.com](https://hireobi.com).
 
 ## Install
 
-Node.js 22 or later is required:
+Node.js 22 or later is required. One line, in a terminal or pasted to your coding agent:
 
 ```bash
-npm install --global @hireobi/cli
-obi --version
-obi skill install
+npm i -g @hireobi/cli && obi
 ```
 
-`obi skill install` registers exactly two personal Skill links, one for each supported local directory, and refuses a conflicting file or broken link, leaving the other Skill intact. `obi skill path` locates the bundled Skill. It also removes the Skill links an earlier alpha of this package registered under its former name, so an assistant lists one Skill. A cloud harness needs its own supported installation path.
+Windows PowerShell 5.1 has no `&&`: run `npm i -g @hireobi/cli`, then `obi`.
 
-## Use
+The first `obi` registers the Obi Skill for Claude Code, Codex, Cursor and OpenCode, and every later command keeps it registered: two personal links, `~/.claude/skills/obi` and `~/.agents/skills/obi`. A conflicting file or link is never replaced: the command still runs, and bare `obi` says what is in the way and how to fix it. `obi skill install` registers it again after a fix and removes the links an earlier alpha of this package made under its former name; `obi skill path` locates the bundled Skill. `OBI_SKILL_LINKS=off` leaves the links to `obi skill install`. A cloud harness needs its own supported installation path.
+
+Run by a coding agent, bare `obi` prints where the Skill's guide is and the next command. In an interactive terminal it shows the Obi mascot and a short welcome, then exits: the conversation happens in the assistant. The greeting is English or French; `OBI_LANGUAGE` overrides `LC_ALL`, `LC_MESSAGES`, then `LANG`.
 
-Open a new assistant chat. In [Claude Code](https://code.claude.com/docs/en/skills) or [Cursor](https://cursor.com/docs/skills), type `/obi`. In Codex CLI, type `$obi` or select Obi through `/skills`; in the desktop app, use the [skills picker](https://learn.chatgpt.com/docs/build-skills). The Skill asks which website to work on and follows the conversation language.
+## Use
 
-Running bare `obi` in an interactive terminal shows the Obi mascot and a short welcome, then exits: the conversation happens in the assistant. The greeting is English or French; `OBI_LANGUAGE` overrides `LC_ALL`, `LC_MESSAGES`, then `LANG`.
+In [Claude Code](https://code.claude.com/docs/en/skills) or [Cursor](https://cursor.com/docs/skills), type `/obi`; in Claude Code, `/reload-skills` lists it when the session started before the install. In Codex CLI, type `$obi` or select Obi through `/skills`; in the desktop app, use the [skills picker](https://learn.chatgpt.com/docs/build-skills). In [OpenCode](https://opencode.ai/docs/skills/), type `@obi`. The Skill asks which website to work on and follows the conversation language.
 
 Run `obi --help` for the complete command grammar and the starting steps. `--json` returns one validated envelope per invocation. After a timeout or a conflict, read the current state before retrying a write; a submission may be resent with the same file and replay key. Credentials stay in the system credential store or an explicitly selected owner-only file, and the CLI never prints tokens. The Skill's references carry the result contracts; `obi skill path` shows where they are installed.
 
@@ -26,7 +26,7 @@ An authorized operator can start an isolated public-company evaluation case with
 
 ## Support
 
-Write to [support@hireobi.com](mailto:support@hireobi.com) or open an issue at [github.com/hireobi/cli](https://github.com/hireobi/cli/issues).
+Run `obi support "<what happened>"`, write to [support@hireobi.com](mailto:support@hireobi.com) or open an issue at [github.com/hireobi/cli](https://github.com/hireobi/cli/issues).
 
 ## Repository development
 

package/dist/obi.js

@@ -14618,6 +14618,7 @@ var optionDefinitions = {
   "field-path": { type: "string" },
   "segment-name": { type: "string" },
   text: { type: "string" },
+  tag: { type: "string" },
   company: { type: "string" },
   "company-unknown": { type: "boolean" },
   product: { type: "string" },
@@ -14632,7 +14633,7 @@ var optionDefinitions = {
   help: { type: "boolean" },
   version: { type: "boolean" }
 };
-var namedCommands = ["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies"];
+var namedCommands = ["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies", "support"];
 var memoryKinds = ["preference", "constraint", "decision"];
 var credentialsActions = ["list", "revoke"];
 var emailSchema = external_exports.email().max(254);
@@ -14651,6 +14652,8 @@ var CliError = class extends Error {
   command;
   authorization;
   account;
+  /** The message `obi support` could not send, with the block to paste. */
+  support;
   constructor(options) {
     super(options.message);
     this.name = "CliError";
@@ -14662,6 +14665,7 @@ var CliError = class extends Error {
     this.command = options.command;
     this.authorization = options.authorization;
     this.account = options.account;
+    this.support = options.support;
   }
 };
 function usageError(message = "Invalid command usage.", options = {}) {
@@ -15174,6 +15178,52 @@ function parseMemoryArguments(positionals, values, context, common2) {
   throw invalid3();
 }
 
+// src/support-contract.ts
+var SUPPORT_EMAIL = "support@hireobi.com";
+var SUPPORT_TAGS = ["market-request", "scope-request"];
+var lastErrorSchema = external_exports.strictObject({
+  code: external_exports.string().regex(/^[A-Z][A-Z0-9_]{0,79}$/),
+  command: external_exports.string().regex(/^[a-z][a-z-]{0,39}$/),
+  at: external_exports.iso.datetime({ offset: true })
+});
+var SupportMessageSchema = external_exports.strictObject({
+  requestId: external_exports.string().regex(/^[0-9a-f]{32}$/),
+  message: external_exports.string().trim().min(1).max(2e3),
+  tag: external_exports.enum(SUPPORT_TAGS).nullable(),
+  diagnostic: external_exports.strictObject({
+    cliVersion: external_exports.string().min(1).max(40),
+    node: external_exports.string().min(1).max(40),
+    os: external_exports.string().min(1).max(120),
+    lastError: lastErrorSchema.nullable()
+  })
+});
+var SupportMessageResponseSchema = external_exports.object({
+  schemaVersion: external_exports.literal(1),
+  data: external_exports.object({ received: external_exports.literal(true) })
+});
+var supportDataSchema = external_exports.object({
+  delivery: external_exports.enum(["sent", "not_sent"]),
+  /** Why nothing was sent: the error code met, null once sent. */
+  reason: external_exports.string().nullable(),
+  supportEmail: external_exports.literal(SUPPORT_EMAIL),
+  mailto: external_exports.string(),
+  tag: external_exports.enum(SUPPORT_TAGS).nullable(),
+  account: external_exports.string().nullable(),
+  origin: external_exports.string().min(1),
+  diagnosticText: external_exports.string()
+});
+
+// src/support-arguments.ts
+function parseSupportArguments(positionals, values, context, common2) {
+  const allowed = /* @__PURE__ */ new Set(["json", "timeout-ms", "tag"]);
+  const message = positionals[1]?.trim() ?? "";
+  const tag = values.tag;
+  if (Object.entries(values).some(([key, value]) => value !== void 0 && value !== false && !allowed.has(key)) || positionals.length !== 2 || message.length === 0 || message.length > 2e3 || tag !== void 0 && !SUPPORT_TAGS.includes(tag)) {
+    throw usageError(`Use support "<message of at most 2000 characters>" [--tag ${SUPPORT_TAGS.join("|")}].`, context);
+  }
+  return { command: "support", supportMessage: message, ...tag === void 0 ? {} : { supportTag: tag }, ...common2 };
+}
+
 // src/research-arguments.ts
 function parseResearchScope(values, context) {
   const locale = values.locale;
@@ -15576,7 +15626,7 @@ function commandMode(command, testServer) {
   if (testServer) {
     return "test";
   }
-  return command === "login" || command === "logout" || command === "status" || command === "credentials" || command === "eval" || command === "harness" || command === "business" || command === "questions" || command === "strategies" || command === "readiness" || command === "memory" ? "product" : "local";
+  return command === "login" || command === "logout" || command === "status" || command === "credentials" || command === "eval" || command === "harness" || command === "business" || command === "questions" || command === "strategies" || command === "readiness" || command === "memory" || command === "support" ? "product" : "local";
 }
 function errorContext(argv) {
   const usesTestServer = argv.some(
@@ -15762,6 +15812,8 @@ function parseCommandArguments(argv) {
     throw usageError("Use research crawl <domain> --checkpoint <file> --output <file> [--batch-size <1-100>] [--locale <tag>] [--pages priority|all] [--max-pages <n|none>] [--retry-failed|--refresh], research capture --input <selected-page-file> --checkpoint <file> --output <corpus-file>, research readiness --input <file> --output <file>, or research video <url> --output <directory> [--frames <0-30>].", context);
   }
   if (positionals[0] === "memory") return parseMemoryArguments(positionals, values, context, { evaluationCaseId, json: json2, timeoutMs });
+  if (positionals[0] === "support") return parseSupportArguments(positionals, values, context, { json: json2, timeoutMs });
+  if (values.tag !== void 0) throw usageError("--tag is only valid with obi support.", context);
   if (stage !== void 0 || input !== void 0 || output !== void 0 || checkpoint !== void 0 || batchSize !== void 0 || retryFailed || refresh || scopeFlagsUsed || values.frames !== void 0 || free || evidenceOptionsUsed) {
     throw usageError("--stage and --input are only valid with obi harness.", context);
   }
@@ -15856,9 +15908,6 @@ function parseCommandArguments(argv) {
   }
   if (command === "doctor") {
     if (evaluationCaseId !== void 0) throw usageError("Doctor cannot use --eval-case.", context);
-    if (timeoutValue !== void 0 && testServer === void 0) {
-      throw usageError("A timeout is only valid with a test server.", context);
-    }
     return { command, json: json2, testServer, timeoutMs };
   }
   if (command === "status") {
@@ -16587,6 +16636,34 @@ var BusinessItemReviewRequestSchema = external_exports.discriminatedUnion("decis
   })
 ]);
 
+// ../shared/lib/geo/markets.ts
+var SUPPORTED_MARKETS = [
+  { iso: "US", name: "United States" },
+  { iso: "CA", name: "Canada" },
+  { iso: "GB", name: "United Kingdom" },
+  { iso: "FR", name: "France" },
+  { iso: "DE", name: "Germany" },
+  { iso: "ES", name: "Spain" },
+  { iso: "IT", name: "Italy" },
+  { iso: "NL", name: "Netherlands" },
+  { iso: "BE", name: "Belgium" },
+  { iso: "CH", name: "Switzerland" },
+  { iso: "AU", name: "Australia" },
+  { iso: "SA", name: "Saudi Arabia" },
+  { iso: "AE", name: "UAE" },
+  { iso: "QA", name: "Qatar" },
+  { iso: "KW", name: "Kuwait" }
+];
+var BY_KEY = new Map(
+  SUPPORTED_MARKETS.flatMap((market) => [
+    [market.iso.toLowerCase(), market],
+    [market.name.toLowerCase(), market]
+  ])
+);
+var SUPPORTED_MARKET_NAMES = SUPPORTED_MARKETS.map(
+  (market) => market.name
+);
+
 // src/product-api-failures.ts
 function transportError(code, message, exitCode, retryable) {
   return new CliError({
@@ -16594,7 +16671,7 @@ function transportError(code, message, exitCode, retryable) {
     message,
     exitCode,
     retryable,
-    nextAction: exitCode === 6 ? "Check the network connection and try again." : "Update the Obi CLI, then try again.",
+    nextAction: exitCode === 6 ? `Check the network connection and try again. If it keeps failing, run obi support "<what happened>" or write to ${SUPPORT_EMAIL} with the command you ran and the error code ${code}.` : "Update the Obi CLI, then try again.",
     mode: "product"
   });
 }
@@ -16624,6 +16701,14 @@ function failureError(failure2, evaluationCaseId) {
     mode: "product",
     nextAction: `Run obi business show${scope} --json and read the current work before preparing a new result.`
   });
+  if (failure2.code === "MARKET_UNSUPPORTED") return new CliError({
+    code: failure2.code,
+    message: "Obi does not measure this market yet.",
+    exitCode: 5,
+    retryable: false,
+    mode: "product",
+    nextAction: `Obi measures in: ${SUPPORTED_MARKET_NAMES.join(", ")}. Choose one of them with the customer, or record the request with obi support "<market and domain>" --tag market-request.`
+  });
   if (failure2.code === "RATE_LIMITED") {
     const wait = failure2.retryAfterSeconds === null ? "a moment" : `${failure2.retryAfterSeconds} seconds`;
     return new CliError({
@@ -17213,7 +17298,7 @@ var originMarker = {
 };
 var helpDataSchema = external_exports.object({
   usage: external_exports.string(),
-  commands: external_exports.array(external_exports.enum(["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies"]))
+  commands: external_exports.array(external_exports.enum(["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies", "support"]))
 });
 var versionDataSchema = external_exports.object({ version: external_exports.string() });
 var skillDataSchema = external_exports.discriminatedUnion("action", [
@@ -17223,6 +17308,13 @@ var skillDataSchema = external_exports.discriminatedUnion("action", [
 var capabilitiesDataSchema = external_exports.object({
   expertise: external_exports.object({ manifestVersion: external_exports.string(), competencies: external_exports.array(external_exports.object({ id: external_exports.string(), reference: external_exports.string(), sha256: external_exports.string(), reviewedOn: external_exports.string(), sources: external_exports.array(external_exports.string()) })) }).optional(),
   version: external_exports.string(),
+  measurementMarkets: external_exports.array(external_exports.object({ iso: external_exports.string().length(2), name: external_exports.string().min(1) })),
+  catalog: external_exports.object({ version: external_exports.string(), currency: external_exports.literal("usd"), monthlyCents: external_exports.object({
+    corePack: external_exports.number().int().positive(),
+    claudePerPack: external_exports.number().int().positive(),
+    perplexityPerPack: external_exports.number().int().positive()
+  }) }),
+  support: external_exports.object({ email: external_exports.literal(SUPPORT_EMAIL) }),
   realLogin: external_exports.literal("available"),
   paidExecution: external_exports.literal("server_authorized"),
   commands: external_exports.object({
@@ -17240,7 +17332,8 @@ var capabilitiesDataSchema = external_exports.object({
     memory: external_exports.object({ actions: external_exports.array(external_exports.string()), options: external_exports.array(external_exports.string()) }).optional(),
     status: external_exports.object({ options: external_exports.array(external_exports.string()) }),
     readiness: external_exports.object({ options: external_exports.array(external_exports.string()) }),
-    strategies: external_exports.object({ options: external_exports.array(external_exports.string()) })
+    strategies: external_exports.object({ options: external_exports.array(external_exports.string()) }),
+    support: external_exports.object({ tags: external_exports.array(external_exports.enum(SUPPORT_TAGS)), options: external_exports.array(external_exports.string()) })
   })
 });
 var localDoctorDataSchema = external_exports.object({
@@ -17249,6 +17342,7 @@ var localDoctorDataSchema = external_exports.object({
   origin: external_exports.string(),
   credentialStorage: external_exports.union([backendKindSchema, external_exports.literal("none")]),
   videoTools: external_exports.object({ ytDlp: external_exports.boolean(), ffmpeg: external_exports.boolean() }),
+  server: external_exports.enum(["reachable", "degraded", "unreachable"]),
   diagnostic: external_exports.literal("completed")
 });
 var loginDataSchema = external_exports.object({
@@ -17545,6 +17639,7 @@ var successfulResultSchema = external_exports.union([
   external_exports.object({ ...envelope2, command: external_exports.literal("harness"), mode: external_exports.literal("product"), data: harnessCoverageDataSchema }),
   ...harnessEvidenceDataSchema.map((schema) => external_exports.object({ ...envelope2, command: external_exports.literal("harness"), mode: external_exports.literal("product"), data: schema })),
   external_exports.object({ ...envelope2, command: external_exports.literal("harness"), mode: external_exports.literal("product"), data: harnessEvidenceExportDataSchema }),
+  external_exports.object({ ...envelope2, command: external_exports.literal("support"), mode: external_exports.literal("product"), data: supportDataSchema }),
   external_exports.object({ ...envelope2, command: external_exports.literal("memory"), mode: external_exports.literal("product"), data: memoryListDataSchema }),
   external_exports.object({ ...envelope2, command: external_exports.literal("memory"), mode: external_exports.literal("product"), data: memoryChangeDataSchema }),
   external_exports.object({
@@ -17595,7 +17690,8 @@ var failedResultSchema = external_exports.object({
   ...originMarker,
   error: errorSchema,
   authorization: pendingAuthorizationViewSchema.optional(),
-  account: confirmedAccountSchema.optional()
+  account: confirmedAccountSchema.optional(),
+  support: supportDataSchema.optional()
 });
 var cliResultSchema = external_exports.union([successfulResultSchema, failedResultSchema]);
 
@@ -18060,6 +18156,9 @@ async function readReadiness(context, accessToken) {
     responseLimitBytes: BUSINESS_REVIEW_RESPONSE_LIMIT_BYTES
   }), ProductReadinessResponseSchema);
 }
+async function sendSupportMessage(context, accessToken, message) {
+  return interpretReview(await send(context, { method: "POST", path: "/api/v1/support", bearer: accessToken, body: SupportMessageSchema.parse(message) }), SupportMessageResponseSchema);
+}
 
 // src/refresh.ts
 import { randomBytes as randomBytes2 } from "node:crypto";
@@ -19210,7 +19309,7 @@ async function submitHarnessObservation(context, path) {
 
 // src/commands.ts
 import { access } from "node:fs/promises";
-import { fileURLToPath } from "node:url";
+import { fileURLToPath as fileURLToPath2 } from "node:url";
 
 // src/onboarding-guide.ts
 var READING_POLL_MS = 6e4;
@@ -20275,6 +20374,111 @@ function memoryText(data) {
 `;
 }
 
+// src/support-commands.ts
+import { createHash as createHash2 } from "node:crypto";
+import { release } from "node:os";
+
+// src/last-error.ts
+import { readFile as readFile6, writeFile as writeFile3 } from "node:fs/promises";
+import { join as join4 } from "node:path";
+var fileIn = (stateDirectory) => join4(stateDirectory, "last-error.json");
+async function recordLastError(stateDirectory, error51) {
+  try {
+    const parsed = lastErrorSchema.safeParse({ ...error51, at: (/* @__PURE__ */ new Date()).toISOString() });
+    if (!parsed.success) return;
+    await ensureStateDirectory(stateDirectory);
+    await writeFile3(fileIn(stateDirectory), JSON.stringify(parsed.data), { mode: 384 });
+  } catch {
+  }
+}
+async function readLastError(stateDirectory) {
+  try {
+    const parsed = lastErrorSchema.safeParse(JSON.parse(await readFile6(fileIn(stateDirectory), "utf8")));
+    return parsed.success ? parsed.data : null;
+  } catch {
+    return null;
+  }
+}
+
+// src/version.ts
+var cliVersion = "0.1.9";
+
+// src/support-commands.ts
+async function probeServer(origin, timeoutMs) {
+  try {
+    const response = await exchange({
+      url: new URL("/api/health", origin.origin),
+      method: "GET",
+      headers: { accept: "application/json" },
+      timeoutMs
+    });
+    return response.status === 200 ? "reachable" : "degraded";
+  } catch {
+    return "unreachable";
+  }
+}
+async function sendSupport(context, input) {
+  const lastError = await readLastError(context.stateDirectory);
+  const diagnostic = { cliVersion, node: process.version, os: `${process.platform} ${release()} ${process.arch}`, lastError };
+  let account2 = null;
+  let reason = null;
+  try {
+    const located = await locateCredential(context);
+    if (located === null) {
+      reason = "AUTH_REQUIRED";
+    } else {
+      account2 = located.credential.account.email;
+      const requestId = createHash2("sha256").update(JSON.stringify([
+        context.origin.origin,
+        account2,
+        input.message,
+        input.tag,
+        (/* @__PURE__ */ new Date()).toISOString().slice(0, 10)
+      ])).digest("hex").slice(0, 32);
+      const { result } = await useCredential(
+        context,
+        located,
+        (token) => sendSupportMessage(context, token, { requestId, ...input, diagnostic })
+      );
+      if (!result.ok) reason = result.failure.httpStatus >= 500 ? "SERVICE_UNAVAILABLE" : result.failure.code;
+    }
+  } catch (error51) {
+    reason = error51 instanceof CliError ? error51.code : "INTERNAL_ERROR";
+  }
+  const diagnosticText = [
+    `Message: ${input.message}`,
+    `Tag: ${input.tag ?? "general"}`,
+    `Obi CLI ${diagnostic.cliVersion} \xB7 Node ${diagnostic.node} \xB7 ${diagnostic.os}`,
+    `Last error: ${lastError === null ? "none" : `${lastError.code} (${lastError.command}) at ${lastError.at}`}`,
+    `Account: ${account2 ?? "not signed in"}`
+  ].join("\n");
+  const data = {
+    delivery: reason === null ? "sent" : "not_sent",
+    reason,
+    supportEmail: SUPPORT_EMAIL,
+    mailto: `mailto:${SUPPORT_EMAIL}`,
+    tag: input.tag,
+    account: account2,
+    origin: context.origin.origin,
+    diagnosticText
+  };
+  if (reason === null) return data;
+  throw new CliError({
+    code: "SUPPORT_NOT_SENT",
+    message: `The message was not sent to Obi support (${reason}).`,
+    exitCode: reason === "AUTH_REQUIRED" ? 3 : 6,
+    retryable: reason !== "AUTH_REQUIRED",
+    mode: "product",
+    command: "support",
+    nextAction: `Write to ${SUPPORT_EMAIL} and paste the block below (support.diagnosticText with --json). Never add a token or a credential file.`,
+    support: data
+  });
+}
+function supportText(data) {
+  return `Sent to Obi support (${data.supportEmail}). The reply goes to ${sanitizeHumanText(data.account ?? "the account email")}.
+`;
+}
+
 // src/scan-output.ts
 function heading(data) {
   const scored = `${data.scoredCells}/${data.expectedCells} answers scored from question version ${data.candidateVersion ?? "none"}.`;
@@ -20399,6 +20603,7 @@ function failureText(result) {
     lines.push(`Approving account: ${sanitizeHumanText(result.account.email)} (${result.account.role}).`);
   }
   lines.push(sanitizeHumanText(result.error.nextAction));
+  if (result.support !== void 0) lines.push("", ...result.support.diagnosticText.split("\n").map(sanitizeHumanText));
   return `${lines.join("\n")}
 `;
 }
@@ -20441,11 +20646,12 @@ function toHumanOutput(result) {
     return `Obi \xB7 AI Search Employee
 
 Get started
-  1. obi skill install
-  2. In a new Claude Code or Cursor chat, type /obi.
+  1. In Claude Code or Cursor, type /obi.
      In Codex CLI, type $obi or choose Obi from /skills.
      In the desktop app, select Obi from the skills picker.
-  3. Give Obi a website or ask to continue saved work.
+     In OpenCode, type @obi.
+  2. Give Obi a website or ask to continue saved work.
+  Every obi command registers the Skill; obi skill install repairs it.
 
 Commands
 ${usage2}
@@ -20457,11 +20663,11 @@ ${usage2}
   }
   if (result.command === "skill") {
     return result.data.action === "path" ? `${sanitizeHumanText(result.data.path)}
-` : `Obi Skill registered locally. In a new Claude Code or Cursor chat, type /obi. In Codex CLI, type $obi or choose Obi from /skills. In the desktop app, select Obi from the skills picker.
+` : `Obi Skill registered locally. In a new Claude Code or Cursor chat, type /obi. In Codex CLI, type $obi or choose Obi from /skills. In the desktop app, select Obi from the skills picker. In OpenCode, type @obi.
 ${result.data.removed.length > 0 ? "Earlier Skill links of this package under another name were removed; use /obi.\n" : ""}`;
   }
   if (result.command === "capabilities") {
-    return "Available: business, capabilities, credentials, doctor, eval, harness, login, logout, memory, questions, readiness, research, skill, status, strategies. Scan execution requires server authorization and a separate customer decision.\n";
+    return "Available: business, capabilities, credentials, doctor, eval, harness, login, logout, memory, questions, readiness, research, skill, status, strategies, support. Scan execution requires server authorization and a separate customer decision.\n";
   }
   if (result.command === "eval") {
     if ("evaluation" in result.data) {
@@ -20476,6 +20682,7 @@ Next: ${sanitizeHumanText(result.data.nextStep)}
     return result.data.action === "list" ? connectionListText(result.data) : revocationText(result.data);
   }
   if (result.command === "memory") return memoryText(result.data);
+  if (result.command === "support") return supportText(result.data);
   if (result.command === "doctor") {
     if (result.mode === "test") {
       return `Synthetic test protocol reachable. State: ${String(result.data.state)}.
@@ -20483,7 +20690,12 @@ Next: ${sanitizeHumanText(result.data.nextStep)}
     }
     const storage = result.data.credentialStorage === "none" ? "no stored connection" : `connection stored in ${storageLabels[result.data.credentialStorage]}`;
     const video = result.data.videoTools.ytDlp ? `video reading available${result.data.videoTools.ffmpeg ? " with frames" : " (captions only; frames need ffmpeg)"}` : "video reading needs yt-dlp";
-    return `Local diagnostic completed for ${sanitizeHumanText(result.data.origin)}: ${storage}; ${video}. No request was sent.
+    const server = {
+      reachable: "Obi server reachable.",
+      degraded: `Obi server degraded: try again later, and write to ${SUPPORT_EMAIL} if it lasts.`,
+      unreachable: `Obi server unreachable: check the network connection, and write to ${SUPPORT_EMAIL} if it lasts.`
+    }[result.data.server];
+    return `Local diagnostic completed for ${sanitizeHumanText(result.data.origin)}: ${storage}; ${video}. ${server}
 `;
   }
   if (result.command === "login") {
@@ -20819,9 +21031,6 @@ async function confirmAccount(account2) {
   }
 }
 
-// src/version.ts
-var cliVersion = "0.1.8";
-
 // src/login.ts
 var loginRequestTimeoutMs = 1e4;
 var maxLinkLength = 200;
@@ -21031,7 +21240,7 @@ function accountMismatch(account2, revoked, expected) {
 }
 async function confirmAndStore(context, store, pair, receivedAt, expectedEmail) {
   const api = context;
-  const release2 = guardNewFamily(api, pair);
+  const release3 = guardNewFamily(api, pair);
   let stored = false;
   try {
     const status = await readAccountStatus(api, pair.accessToken);
@@ -21062,7 +21271,7 @@ async function confirmAndStore(context, store, pair, receivedAt, expectedEmail)
       }
       await store.writeCredential(credentialFromPair(context.origin.origin, pair, confirmed, receivedAt));
       stored = true;
-      release2();
+      release3();
     });
     try {
       await store.removePending();
@@ -21082,7 +21291,7 @@ async function confirmAndStore(context, store, pair, receivedAt, expectedEmail)
     }
     throw error51;
   } finally {
-    release2();
+    release3();
   }
 }
 async function login(session, options) {
@@ -21221,12 +21430,12 @@ async function proceedToFirstScan(context, candidateVersion) {
 }
 
 // src/research-commands.ts
-import { readFile as readFile6, stat as stat6 } from "node:fs/promises";
+import { readFile as readFile7, stat as stat6 } from "node:fs/promises";
 import { isDeepStrictEqual as isDeepStrictEqual2 } from "node:util";
 import { resolve as resolve2 } from "node:path";
 
 // ../geo-engine/src/geo/host-acquisition.ts
-import { createHash as createHash5 } from "node:crypto";
+import { createHash as createHash6 } from "node:crypto";
 
 // ../geo-engine/src/url-safety.ts
 var PRIVATE_IP_RE = /^(127\.|10\.|192\.168\.|169\.254\.|172\.(1[6-9]|2\d|3[01])\.|0\.)/;
@@ -21397,7 +21606,7 @@ async function safeFetch(url3, opts = {}) {
     if (!REDIRECT_STATUSES.has(res.status)) return res;
     const location = res.headers.get("location");
     if (!location) return res;
-    await release(res);
+    await release2(res);
     let next;
     try {
       next = new URL(location, current).href;
@@ -21408,7 +21617,7 @@ async function safeFetch(url3, opts = {}) {
   }
   throw new TooManyRedirectsError(url3, maxRedirects);
 }
-async function release(res) {
+async function release2(res) {
   await res.body?.cancel().catch((err) => {
     console.warn("[safe-fetch] cancelling an unread body failed:", err);
   });
@@ -21416,7 +21625,7 @@ async function release(res) {
 async function readCapped(res, maxBytes = MAX_RESPONSE_BYTES) {
   const declared = Number(res.headers.get("content-length"));
   if (Number.isFinite(declared) && declared > maxBytes) {
-    await release(res);
+    await release2(res);
     return "";
   }
   if (!res.body) return (await res.text()).slice(0, maxBytes);
@@ -21582,7 +21791,7 @@ function createVettedTransport(options = {}) {
 }
 
 // ../geo-engine/src/geo/page-evidence.ts
-import { createHash as createHash2 } from "node:crypto";
+import { createHash as createHash3 } from "node:crypto";
 
 // ../node_modules/parse5/dist/common/unicode.js
 var UNDEFINED_CODE_POINTS = /* @__PURE__ */ new Set([
@@ -29502,7 +29711,7 @@ function* descendants(node) {
   }
 }
 function sha256(s) {
-  return createHash2("sha256").update(s).digest("hex");
+  return createHash3("sha256").update(s).digest("hex");
 }
 var SKIPPED = /* @__PURE__ */ new Set(["script", "style", "noscript", "template", "svg", "iframe", "object", "canvas", "select", "head"]);
 var HEADING = /^h([1-6])$/;
@@ -29793,7 +30002,7 @@ function headingTexts(html, level) {
 }
 
 // ../geo-engine/src/geo/selected-page-evidence.ts
-import { createHash as createHash3 } from "node:crypto";
+import { createHash as createHash4 } from "node:crypto";
 
 // ../geo-engine/src/geo/material-fact-contract.ts
 var text = external_exports.string().trim().min(1);
@@ -29899,10 +30108,10 @@ var SelectedPageEvidenceSchema = external_exports.strictObject({
 function selectedPageId(pageUrl) {
   const selected = new URL(pageUrl);
   selected.hash = "";
-  return `page:${createHash3("sha256").update(selected.href).digest("hex")}`;
+  return `page:${createHash4("sha256").update(selected.href).digest("hex")}`;
 }
 function observationBodyHash(body) {
-  return createHash3("sha256").update(body).digest("hex");
+  return createHash4("sha256").update(body).digest("hex");
 }
 function sameSelectedHost(selectedUrl, finalUrl) {
   const host = (value) => new URL(value).hostname.toLowerCase().replace(/^www\./, "");
@@ -30354,14 +30563,14 @@ function evaluateAccess(parsed, token, url3, fallbackTokens = []) {
 }
 
 // ../geo-engine/src/geo/readiness/findings.ts
-import { createHash as createHash4 } from "node:crypto";
+import { createHash as createHash5 } from "node:crypto";
 function issueId(origin, checkId, scopeKey) {
-  return createHash4("sha256").update(`${origin.toLowerCase()}
+  return createHash5("sha256").update(`${origin.toLowerCase()}
 ${checkId}
 ${scopeKey}`).digest("hex").slice(0, 16);
 }
 function fingerprint(body) {
-  return body === null ? null : createHash4("sha256").update(body).digest("hex");
+  return body === null ? null : createHash5("sha256").update(body).digest("hex");
 }
 function makeFinding({ origin, scopeKey, policyChoice = false, materiality = "site", ...rest }) {
   return { id: issueId(origin, rest.checkId, scopeKey), policyChoice, materiality, ...rest };
@@ -30864,7 +31073,7 @@ function validateState(raw) {
   return state;
 }
 function matchesFingerprint(observation) {
-  return observation.body === null ? observation.bodyHash === null : observation.bodyHash === createHash5("sha256").update(observation.body).digest("hex");
+  return observation.body === null ? observation.bodyHash === null : observation.bodyHash === createHash6("sha256").update(observation.body).digest("hex");
 }
 function pendingProbeRequests(state) {
   const home = state.resources.find((resource) => resource.kind === "page" && resource.url === `${state.origin}/`);
@@ -32635,7 +32844,7 @@ async function readJson(path, maxBytes = maxCorpusBytes) {
   try {
     const info = await stat6(path);
     if (!info.isFile() || info.size > maxBytes) throw fileError(`The research file is not a regular JSON file within the ${maxBytes / 1024 / 1024} MiB local limit.`);
-    const content = await readFile6(path);
+    const content = await readFile7(path);
     if (content.byteLength > maxBytes) throw fileError(`The research file exceeds the ${maxBytes / 1024 / 1024} MiB local limit.`);
     return JSON.parse(content.toString("utf8"));
   } catch (error51) {
@@ -32823,8 +33032,8 @@ async function captureResearch(input, checkpoint, output) {
 
 // src/research-video.ts
 import { spawn as spawn2 } from "node:child_process";
-import { mkdir as mkdir2, readdir as readdir3, readFile as readFile7, rm as rm2 } from "node:fs/promises";
-import { join as join4, resolve as resolve3 } from "node:path";
+import { mkdir as mkdir2, readdir as readdir3, readFile as readFile8, rm as rm2 } from "node:fs/promises";
+import { join as join5, resolve as resolve3 } from "node:path";
 var maxToolOutputBytes = 256 * 1024;
 var runTool = (executable, args, timeoutMs) => new Promise((done, fail) => {
   const child = spawn2(executable, [...args], { stdio: ["ignore", "pipe", "pipe"], shell: false });
@@ -32923,12 +33132,12 @@ async function readVideo(rawUrl, outputDir, frames, runner = runTool) {
   const directory = resolve3(outputDir);
   await mkdir2(directory, { recursive: true });
   const base = ["--ignore-config", "--no-playlist", "--quiet", "--no-warnings"];
-  const target = ["-o", join4(templateSafe(directory), "video.%(ext)s")];
+  const target = ["-o", join5(templateSafe(directory), "video.%(ext)s")];
   const failed = (message, nextAction) => videoError("RESEARCH_VIDEO_FAILED", message, nextAction);
   const metadata = await runner("yt-dlp", [...base, "--skip-download", "--write-info-json", ...target, url3.href], 18e4).catch((error51) => {
     throw error51.code === "ENOENT" ? missingTool(frames) : failed("yt-dlp did not finish reading this video page.", "Retry once; a slow or blocked network leaves the video unread.");
   });
-  const info = metadata.exitCode === 0 ? JSON.parse(await readFile7(join4(directory, "video.info.json"), "utf8")) : null;
+  const info = metadata.exitCode === 0 ? JSON.parse(await readFile8(join5(directory, "video.info.json"), "utf8")) : null;
   if (!info) {
     throw failed("yt-dlp could not read this video page.", "Check the address in a browser; a private or region-locked video stays unread.");
   }
@@ -32950,11 +33159,11 @@ async function readVideo(rawUrl, outputDir, frames, runner = runTool) {
     if (subtitles?.exitCode !== 0 || !file2) {
       throw failed("The video has captions but their download failed.", "Retry in a minute; the platform may be rate-limiting.");
     }
-    segments = parseVtt(await readFile7(join4(directory, file2), "utf8"));
+    segments = parseVtt(await readFile8(join5(directory, file2), "utf8"));
   }
   const duration3 = typeof info.duration === "number" ? info.duration : null;
   const frameFiles = [];
-  const output = join4(directory, "transcript.json");
+  const output = join5(directory, "transcript.json");
   const save = () => writeJsonAtomic(
     output,
     {
@@ -32974,7 +33183,7 @@ async function readVideo(rawUrl, outputDir, frames, runner = runTool) {
     throw failed("The video's length is unknown, so frames cannot be spaced.", "Retry with --frames 0 to keep the captions only.");
   }
   if (frames > 0 && duration3) {
-    const removeMedia = async () => Promise.all((await readdir3(directory)).filter((file2) => file2.startsWith("media.")).map((file2) => rm2(join4(directory, file2), { force: true })));
+    const removeMedia = async () => Promise.all((await readdir3(directory)).filter((file2) => file2.startsWith("media.")).map((file2) => rm2(join5(directory, file2), { force: true })));
     await removeMedia();
     try {
       const download = await runner("yt-dlp", [
@@ -32982,7 +33191,7 @@ async function readVideo(rawUrl, outputDir, frames, runner = runTool) {
         "-f",
         "b[height<=480]/bv*[height<=480]/w",
         "-o",
-        join4(templateSafe(directory), "media.%(ext)s"),
+        join5(templateSafe(directory), "media.%(ext)s"),
         url3.href
       ], 6e5).catch(() => {
         throw failed("The video download did not finish.", "Retry with --frames 0 to keep the captions only.");
@@ -33002,14 +33211,14 @@ async function readVideo(rawUrl, outputDir, frames, runner = runTool) {
           "-ss",
           String(atSeconds),
           "-i",
-          join4(directory, media),
+          join5(directory, media),
           "-frames:v",
           "1",
           "-vf",
           "scale=1024:-2",
-          join4(directory, file2)
+          join5(directory, file2)
         ], 6e4).catch(() => null);
-        if (extracted?.exitCode === 0) frameFiles.push({ file: join4(directory, file2), atSeconds });
+        if (extracted?.exitCode === 0) frameFiles.push({ file: join5(directory, file2), atSeconds });
       }
     } finally {
       await removeMedia();
@@ -33128,13 +33337,15 @@ async function showAlphaStrategies(context) {
 // src/skill-install.ts
 import { lstat as lstat3, mkdir as mkdir3, readlink, realpath, symlink, unlink as unlink5 } from "node:fs/promises";
 import { homedir as homedir2 } from "node:os";
-import { join as join5, sep } from "node:path";
+import { join as join6, sep } from "node:path";
+import { fileURLToPath } from "node:url";
 var skillDirectories = [".agents/skills", ".claude/skills"];
 var alphaSkillName = "heysurface";
+var messageLimit = 300;
 function installError(message) {
   return new CliError({
     code: "SKILL_INSTALL_FAILED",
-    message,
+    message: message.length > messageLimit ? `${message.slice(0, messageLimit - 1)}\u2026` : message,
     exitCode: 4,
     retryable: false,
     nextAction: "Check the personal Skill directories and run obi skill install again. Existing files are never replaced.",
@@ -33164,7 +33375,7 @@ async function linksIntoPackage(path, packageRoot) {
     return target === packageRoot || target.startsWith(`${packageRoot}${sep}`);
   } catch {
     try {
-      return await readlink(path) === join5(packageRoot, "skills", alphaSkillName);
+      return await readlink(path) === join6(packageRoot, "skills", alphaSkillName);
     } catch {
       return false;
     }
@@ -33173,37 +33384,70 @@ async function linksIntoPackage(path, packageRoot) {
 async function installSkill(sourceDirectory) {
   const source = await realpath(sourceDirectory);
   const home = homedir2();
-  const destinations = skillDirectories.map((directory) => join5(home, directory, "obi"));
+  const destinations = skillDirectories.map((directory) => join6(home, directory, "obi"));
+  const registered = async (destination) => {
+    try {
+      return await realpath(destination) === source;
+    } catch {
+      return false;
+    }
+  };
   for (const destination of destinations) {
-    if (!await existingPath(destination)) continue;
+    if (!await existingPath(destination) || await registered(destination)) continue;
+    let target = null;
     try {
-      if (await realpath(destination) === source) continue;
+      if ((await lstat3(destination)).isSymbolicLink()) target = await readlink(destination);
     } catch {
+      continue;
     }
-    throw installError(`A Skill entry already exists at ${destination}.`);
+    throw installError(target === null ? `A Skill entry already exists: a file or folder this package did not create. Move it, then run obi skill install. It is at ${destination}` : `A Skill entry already exists: a link outside this installation. To use this installation, remove that link, then run obi skill install. The link: ${destination} -> ${target}`);
   }
   for (const destination of destinations) {
-    if (await existingPath(destination)) continue;
+    if (await registered(destination)) continue;
     try {
-      await mkdir3(join5(destination, ".."), { recursive: true });
+      await mkdir3(join6(destination, ".."), { recursive: true });
       await symlink(source, destination, process.platform === "win32" ? "junction" : "dir");
     } catch {
-      throw installError(`Could not register the Obi Skill at ${destination}.`);
+      if (await registered(destination)) continue;
+      throw installError(`Could not register the Obi Skill. Check that its folder can be written, then run obi skill install. The link to make: ${destination}`);
     }
   }
   const removed = [];
-  const packageRoot = await realpath(join5(source, "..", ".."));
-  for (const alpha of skillDirectories.map((directory) => join5(home, directory, alphaSkillName))) {
+  const packageRoot = await realpath(join6(source, "..", ".."));
+  for (const alpha of skillDirectories.map((directory) => join6(home, directory, alphaSkillName))) {
     if (!await linksIntoPackage(alpha, packageRoot)) continue;
     try {
       await unlink5(alpha);
-    } catch {
+    } catch (error51) {
+      if (error51.code === "ENOENT") continue;
       throw installError(`Could not retire the earlier Skill link at ${alpha}.`);
     }
     removed.push(alpha);
   }
   return { paths: destinations, removed };
 }
+function packagedSkillDirectory() {
+  return fileURLToPath(new URL("../skills/obi", import.meta.url));
+}
+async function registerSkillLinks(sourceDirectory, env) {
+  if (env.OBI_SKILL_LINKS === "off") return { state: "off" };
+  try {
+    const source = await realpath(sourceDirectory);
+    let present3 = true;
+    for (const directory of skillDirectories) {
+      try {
+        if (await realpath(join6(homedir2(), directory, "obi")) !== source) present3 = false;
+      } catch {
+        present3 = false;
+      }
+    }
+    if (present3) return { state: "present" };
+    await installSkill(sourceDirectory);
+    return { state: "created" };
+  } catch (error51) {
+    return { state: "failed", reason: error51 instanceof CliError ? error51.message : "The Obi Skill could not be registered." };
+  }
+}
 
 // src/transport.ts
 var loopbackOrigin = /^http:\/\/(127\.0\.0\.1|\[::1\]):([1-9][0-9]{0,4})$/;
@@ -33328,7 +33572,7 @@ async function fetchTestStatus(origin, timeoutMs) {
 }
 
 // src/commands.ts
-var usage = "obi <business|capabilities|credentials|doctor|eval|harness|login|logout|memory|questions|readiness|research|skill|status|strategies> [--json]\n       obi eval <start <domain>|list> [--json]\n       Evaluation commands: add --eval-case <id> to status, business, harness, readiness, questions and strategies\n       obi harness coverage [--json]\n       obi harness activate --free [--json]\n       obi harness show --stage <business|questions|strategies> [--packs <1-40> for questions review] [--json]\n       obi harness submit --stage <business|questions|strategies> --input <file> [--json]\n       obi harness outcomes --task-id <UUID> [--cursor <event-id>] [--limit <1-50>] [--json]\n       obi harness observe --input <file> [--json]\n       obi harness purchase <quote|prepare|status|resume|change> [--input <quote-and-basket.json>; resume: quote ID only] [--confirm-checkout for prepare] [--confirm-change for change] [--json]\n       obi harness artifact <save|show|history|export> --work-id <UUID> [--input <JSON file>] [--revision <n>] [--output <new file>] [--cursor <opaque>] [--limit <1-100>] [--json]\n       obi harness evidence --work-id <UUID> [--kind claims|sections|measurements] [--cursor <opaque>] [--query <text>] [--url <exact URL>] [--all --output <file>] [--json]\n       obi research crawl <domain> --checkpoint <file> --output <file> [--batch-size <1-100>] [--locale <tag>] [--pages priority|all] [--max-pages <n|none>] [--retry-failed|--refresh] [--json]\n       obi research capture --input <selected-page-file> --checkpoint <file> --output <corpus-file> [--json]\n       obi research readiness --input <corpus-file> --output <report-file> [--json]\n       obi research video <url> --output <directory> [--frames <0-30>] [--json]\n       obi memory <list|add --kind <preference|constraint|decision> --text <words>|remove <id>> [--eval-case <id>] [--json]\n       obi questions show [--json]\n       obi questions prepare --confirm-preparation [--json]\n       obi questions edit <id> --candidate-version <n> --text <wording> [--json]\n       obi questions proceed --candidate-version <n> --confirm-scan [--json]\n       obi questions results [--run latest|first|<run ID>] [--from <question position>] [--json]\n       obi readiness [--json]\n       obi strategies [--json]\n       obi skill <path|install> [--json]\n       obi credentials <list|revoke <credentialId>> [--force]\n       obi business start --confirm-domain <domain>\n       obi business show\n       obi business discover --snapshot <token> --confirm-discovery\n       obi business clarify-identity --snapshot <token> --company <name>\n       obi business clarify-fact <itemId> --snapshot <token> --field-path <path> --text <answer> --company <current name> --company-wide [--segment-name <name> for segments.append]\n       obi business challenge <itemId> --snapshot <token> --reason <reason>\n       obi business correct <itemId> --snapshot <token> --claim-id <id> --field-path <path> --text <affirmative correction> --company <current name> --company-wide\n       obi business remove <competitorItemId> --snapshot <token> [--reason <reason>]\n       obi business <add-competitor|remove-competitor|readd-competitor> <domain> --snapshot <token> [--reason <reason>]\n       obi business confirm --snapshot <token> --confirm-review\n       Global options: --json, --timeout-ms <100-30000>; login: --expect-email <email> [--storage file]; status/doctor test: --test-server <origin>";
+var usage = 'obi <business|capabilities|credentials|doctor|eval|harness|login|logout|memory|questions|readiness|research|skill|status|strategies|support> [--json]\n       obi support "<message>" [--tag market-request|scope-request] [--json]\n       obi eval <start <domain>|list> [--json]\n       Evaluation commands: add --eval-case <id> to status, business, harness, readiness, questions and strategies\n       obi harness coverage [--json]\n       obi harness activate --free [--json]\n       obi harness show --stage <business|questions|strategies> [--packs <1-40> for questions review] [--json]\n       obi harness submit --stage <business|questions|strategies> --input <file> [--json]\n       obi harness outcomes --task-id <UUID> [--cursor <event-id>] [--limit <1-50>] [--json]\n       obi harness observe --input <file> [--json]\n       obi harness purchase <quote|prepare|status|resume|change> [--input <quote-and-basket.json>; resume: quote ID only] [--confirm-checkout for prepare] [--confirm-change for change] [--json]\n       obi harness artifact <save|show|history|export> --work-id <UUID> [--input <JSON file>] [--revision <n>] [--output <new file>] [--cursor <opaque>] [--limit <1-100>] [--json]\n       obi harness evidence --work-id <UUID> [--kind claims|sections|measurements] [--cursor <opaque>] [--query <text>] [--url <exact URL>] [--all --output <file>] [--json]\n       obi research crawl <domain> --checkpoint <file> --output <file> [--batch-size <1-100>] [--locale <tag>] [--pages priority|all] [--max-pages <n|none>] [--retry-failed|--refresh] [--json]\n       obi research capture --input <selected-page-file> --checkpoint <file> --output <corpus-file> [--json]\n       obi research readiness --input <corpus-file> --output <report-file> [--json]\n       obi research video <url> --output <directory> [--frames <0-30>] [--json]\n       obi memory <list|add --kind <preference|constraint|decision> --text <words>|remove <id>> [--eval-case <id>] [--json]\n       obi questions show [--json]\n       obi questions prepare --confirm-preparation [--json]\n       obi questions edit <id> --candidate-version <n> --text <wording> [--json]\n       obi questions proceed --candidate-version <n> --confirm-scan [--json]\n       obi questions results [--run latest|first|<run ID>] [--from <question position>] [--json]\n       obi readiness [--json]\n       obi strategies [--json]\n       obi skill <path|install> [--json]\n       obi credentials <list|revoke <credentialId>> [--force]\n       obi business start --confirm-domain <domain>\n       obi business show\n       obi business discover --snapshot <token> --confirm-discovery\n       obi business clarify-identity --snapshot <token> --company <name>\n       obi business clarify-fact <itemId> --snapshot <token> --field-path <path> --text <answer> --company <current name> --company-wide [--segment-name <name> for segments.append]\n       obi business challenge <itemId> --snapshot <token> --reason <reason>\n       obi business correct <itemId> --snapshot <token> --claim-id <id> --field-path <path> --text <affirmative correction> --company <current name> --company-wide\n       obi business remove <competitorItemId> --snapshot <token> [--reason <reason>]\n       obi business <add-competitor|remove-competitor|readd-competitor> <domain> --snapshot <token> [--reason <reason>]\n       obi business confirm --snapshot <token> --confirm-review\n       Global options: --json, --timeout-ms <100-30000>; login: --expect-email <email> [--storage file]; status/doctor test: --test-server <origin>';
 function success2(command, mode, data) {
   return cliResultSchema.parse({ schemaVersion: 1, command, mode, ok: true, data });
 }
@@ -33355,7 +33599,7 @@ function statusResult(status) {
   }
   return success2("status", "test", status);
 }
-async function localDoctor(runtime) {
+async function localDoctor(runtime, timeoutMs) {
   const located = await locateCredential(runtime);
   return success2("doctor", "local", {
     node: process.version,
@@ -33363,6 +33607,7 @@ async function localDoctor(runtime) {
     origin: runtime.origin.origin,
     credentialStorage: located === null ? "none" : located.store.backend.kind,
     videoTools: await videoTools(),
+    server: await probeServer(runtime.origin, timeoutMs),
     diagnostic: "completed"
   });
 }
@@ -33375,7 +33620,7 @@ async function executeCommand(args, runtime) {
   if (args.command === "help") {
     return success2("help", "local", {
       usage,
-      commands: ["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies"]
+      commands: ["business", "capabilities", "credentials", "doctor", "eval", "harness", "login", "logout", "memory", "questions", "readiness", "research", "skill", "status", "strategies", "support"]
     });
   }
   if (args.command === "version") {
@@ -33384,6 +33629,13 @@ async function executeCommand(args, runtime) {
   if (args.command === "capabilities") {
     return success2("capabilities", "local", {
       version: cliVersion,
+      measurementMarkets: SUPPORTED_MARKETS,
+      catalog: { version: CLI_DOLLAR_CATALOG_VERSION, currency: "usd", monthlyCents: {
+        corePack: CLI_DOLLAR_PRICES.core,
+        claudePerPack: CLI_DOLLAR_PRICES.claude,
+        perplexityPerPack: CLI_DOLLAR_PRICES.perplexity
+      } },
+      support: { email: SUPPORT_EMAIL },
       realLogin: "available",
       paidExecution: "server_authorized",
       expertise: { manifestVersion: AI_SEARCH_COMPETENCY_VERSION, competencies: AI_SEARCH_COMPETENCIES },
@@ -33402,13 +33654,14 @@ async function executeCommand(args, runtime) {
         skill: { actions: ["path", "install"], options: ["--json"] },
         status: { options: ["--json", "--eval-case", "--test-server", "--timeout-ms"] },
         readiness: { options: ["--json", "--eval-case", "--timeout-ms"] },
-        strategies: { options: ["--json", "--eval-case", "--timeout-ms"] }
+        strategies: { options: ["--json", "--eval-case", "--timeout-ms"] },
+        support: { tags: [...SUPPORT_TAGS], options: ["--json", "--tag", "--timeout-ms"] }
       }
     });
   }
   if (args.command === "doctor") {
     if (args.testServer === void 0) {
-      return localDoctor(runtime);
+      return localDoctor(runtime, args.timeoutMs);
     }
     const status = await fetchTestStatus(args.testServer, args.timeoutMs);
     return success2("doctor", "test", {
@@ -33418,7 +33671,7 @@ async function executeCommand(args, runtime) {
     });
   }
   if (args.command === "skill") {
-    const path = fileURLToPath(new URL("../skills/obi/SKILL.md", import.meta.url));
+    const path = fileURLToPath2(new URL("../skills/obi/SKILL.md", import.meta.url));
     try {
       await access(path);
     } catch {
@@ -33431,7 +33684,7 @@ async function executeCommand(args, runtime) {
         mode: "local"
       });
     }
-    return success2("skill", "local", args.skillAction === "install" ? { action: "install", ...await installSkill(fileURLToPath(new URL("../skills/obi", import.meta.url))) } : { action: "path", path });
+    return success2("skill", "local", args.skillAction === "install" ? { action: "install", ...await installSkill(fileURLToPath2(new URL("../skills/obi", import.meta.url))) } : { action: "path", path });
   }
   if (args.command === "login") {
     return success2("login", "product", await login(session, {
@@ -33488,6 +33741,9 @@ async function executeCommand(args, runtime) {
       }
     ) : args.researchAction === "video" ? await readVideo(args.researchVideoUrl, args.researchOutput, args.researchFrames ?? 0) : args.researchAction === "capture" ? await captureResearch(args.researchInput, args.researchCheckpoint, args.researchOutput) : await assessResearch(args.researchInput, args.researchOutput));
   }
+  if (args.command === "support") {
+    return success2("support", "product", await sendSupport(session, { message: args.supportMessage, tag: args.supportTag ?? null }));
+  }
   if (args.command === "memory") {
     if (args.memoryAction === "list") return success2("memory", "product", await listMemory(session));
     return success2("memory", "product", await changeMemory(session, args.memoryAction === "add" ? { op: "add", kind: args.memoryKind, text: args.memoryText } : { op: "remove", id: args.memoryEntryId }));
@@ -33540,23 +33796,51 @@ async function executeCommand(args, runtime) {
   return success2("status", "product", await readStatus(session));
 }
 
+// src/start.ts
+import { join as join7 } from "node:path";
+var linkStates = {
+  created: "created",
+  present: "present",
+  off: "off (OBI_SKILL_LINKS), run obi skill install"
+};
+function renderStart(input) {
+  const links = input.registration.state === "failed" ? `failed. ${sanitizeHumanText(input.registration.reason)}` : `${linkStates[input.registration.state]}.`;
+  const account2 = input.accountEmail === null ? [
+    "Account: not connected. Connect: obi login --expect-email <the user's email> (prints an address and a code for the user).",
+    "Then: obi status --json"
+  ] : [`Account: connected as ${sanitizeHumanText(input.accountEmail)}. Its state: obi status --json`];
+  return [
+    `Obi CLI ${cliVersion}, npm package @hireobi/cli. Skill links: ${links}`,
+    `Agent guide, a local file in this package: ${sanitizeHumanText(join7(input.skillDirectory, "SKILL.md"))}`,
+    ...account2,
+    input.registration.state === "created" || input.registration.state === "present" ? "In the agent: /obi (Claude Code, Cursor), $obi (Codex) or @obi (OpenCode); /reload-skills in Claude Code if /obi is missing." : "In the agent: /obi, $obi and @obi exist once the Skill links do; until then, read the guide above.",
+    'The obi command comes only from npm i -g @hireobi/cli; the npm package named "obi" is unrelated.'
+  ].join("\n") + "\n";
+}
+
 // src/welcome.ts
 var copy = {
   en: {
     tagline: "Your AI Search employee.",
     greeting: "Hi, I'm Obi. Let's get you discovered.",
-    install: "Get started: obi skill install",
+    ready: "I'm in your assistants: Claude Code, Codex, Cursor and OpenCode.",
+    off: "Get started: obi skill install",
+    failed: "I could not join your assistants.",
     claude: "Claude Code / Cursor: /obi",
     codex: "Codex: $obi or the skills picker",
+    opencode: "OpenCode: @obi",
     next: "Then give me your website in your assistant's chat.",
     help: "All commands: obi --help"
   },
   fr: {
     tagline: "Ton agent AI Search.",
     greeting: "Salut, moi c'est Obi. Faisons conna\xEEtre ton entreprise.",
-    install: "Pour d\xE9marrer : obi skill install",
+    ready: "Je suis dans tes assistants : Claude Code, Codex, Cursor et OpenCode.",
+    off: "Pour d\xE9marrer : obi skill install",
+    failed: "Je n'ai pas pu rejoindre tes assistants.",
     claude: "Claude Code / Cursor : /obi",
     codex: "Codex : $obi ou le s\xE9lecteur de skills",
+    opencode: "OpenCode : @obi",
     next: "Donne-moi ensuite ton site dans ton assistant.",
     help: "Toutes les commandes : obi --help"
   }
@@ -33610,7 +33894,7 @@ function mascotLine(row, depth) {
     return `${color(top, false, depth)}${color(bottom, true, depth)}\u2580`;
   }).join("") + reset;
 }
-function renderWelcome(env, columns, colorDepth) {
+function renderWelcome(env, columns, colorDepth, registration) {
   const locale = [env.OBI_LANGUAGE, env.LC_ALL, env.LC_MESSAGES, env.LANG].find((value) => value?.trim())?.trim().toLowerCase() ?? "en";
   const text2 = copy[/^fr(?:[-_.@]|$)/.test(locale) ? "fr" : "en"];
   const colorful = columns >= 64 && colorDepth >= 8 && env.TERM !== "dumb" && !env.NO_COLOR;
@@ -33619,14 +33903,16 @@ function renderWelcome(env, columns, colorDepth) {
     return `  ${mascotLine(row, colorDepth)}     ${aside}`;
   }).join("\n") : `Obi \xB7 ${cliVersion}
 ${text2.tagline}`;
+  const skill = registration.state === "failed" ? `${text2.failed} ${sanitizeHumanText(registration.reason)}` : registration.state === "off" ? text2.off : text2.ready;
   return `
 ${title}
 
 ${text2.greeting}
 
-  ${text2.install}
+  ${skill}
   ${text2.claude}
   ${text2.codex}
+  ${text2.opencode}
 
 ${text2.next}
 ${text2.help}
@@ -33648,7 +33934,8 @@ function failureResult(error51, command, mode) {
       nextAction: error51.nextAction
     },
     authorization: error51.authorization,
-    account: error51.account
+    account: error51.account,
+    support: error51.support
   });
 }
 function markOrigin(result, origin) {
@@ -33658,6 +33945,8 @@ async function runCli() {
   const argv = process.argv.slice(2);
   const json2 = wantsJson(argv);
   const welcome = argv.length === 0 && process.stdin.isTTY === true && process.stdout.isTTY === true;
+  const start = argv.length === 0 && !welcome;
+  const registration = await registerSkillLinks(packagedSkillDirectory(), process.env);
   let command = "unknown";
   let mode = commandMode(
     void 0,
@@ -33675,14 +33964,21 @@ async function runCli() {
     return resolved;
   };
   try {
-    const args = parseCommandArguments(welcome ? ["--help"] : argv);
+    const args = parseCommandArguments(welcome || start ? ["--help"] : argv);
     command = args.command;
     mode = commandMode(args.command, args.testServer !== void 0);
     origin = activeOrigin();
-    const result = await executeCommand(args, {
-      origin,
-      stateDirectory: await prepareStateDirectory(process.env)
-    });
+    const runtime = { origin, stateDirectory: await prepareStateDirectory(process.env) };
+    if (start) {
+      const located = await locateCredential(runtime, "skip");
+      process.stdout.write(renderStart({
+        registration,
+        skillDirectory: packagedSkillDirectory(),
+        accountEmail: located === null ? null : located.credential.account.email
+      }));
+      return;
+    }
+    const result = await executeCommand(args, runtime);
     const evaluationCaseId = args.evaluationCaseId;
     const scopeNext = (data) => ({
       ...data,
@@ -33694,7 +33990,7 @@ async function runCli() {
       ...result.ok && result.command === "status" && result.mode === "product" ? { data: scopeNext(result.data) } : result.ok && result.command === "business" && result.data.action === "start" ? { data: scopeNext(result.data) } : {}
     }) : result;
     if (scoped.ok && welcome) {
-      process.stdout.write(renderWelcome(process.env, process.stdout.columns, process.stdout.getColorDepth()));
+      process.stdout.write(renderWelcome(process.env, process.stdout.columns, process.stdout.getColorDepth(), registration));
     } else {
       writeResult(markOrigin(scoped, origin), args.json);
     }
@@ -33712,7 +34008,14 @@ async function runCli() {
       retryable: false,
       nextAction: "Run the command again or inspect the local CLI installation."
     });
-    writeResult(markOrigin(failureResult(knownError, knownError.command ?? command, mode), origin), json2);
+    const failure2 = failureResult(knownError, knownError.command ?? command, mode);
+    if (failure2.mode !== "test" && failure2.command !== "support" && knownError.code !== "AUTHORIZATION_PENDING") {
+      await prepareStateDirectory(process.env).then((directory) => recordLastError(
+        directory,
+        { code: knownError.code, command: failure2.command }
+      ), () => void 0);
+    }
+    writeResult(markOrigin(failure2, origin), json2);
     process.exitCode = knownError.exitCode;
   } finally {
     closeIdleConnections();

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@hireobi/cli",
-  "version": "0.1.8",
+  "version": "0.1.9",
   "description": "Obi, the AI Search Employee: CLI and Skill for your AI assistant",
   "keywords": [
     "obi",

package/skills/obi/SKILL.md

@@ -33,7 +33,7 @@ Obi remembers what the customer tells it to remember, across hosts and sessions:
 
 Check `obi capabilities --json`, then `obi memory list --json`. Keep the account context established by the conversation or CLI result. For the ordinary connected business, use `obi status --json` and work commands without `--eval-case`; do not look for an evaluation ID. Only an explicitly established public-company evaluation uses `--eval-case <id>` on every account or work read and write, including `status`. `obi eval list --json` recovers a missing ID for that already selected evaluation; a missing ID alone never establishes one. If the machine is unconnected, follow [connection](references/connection.md). If the customer names a public company to evaluate, use its domain only after they choose it inside this conversation. An eligible operator starts an isolated case with `obi eval start <domain> --json`. Never use the ordinary account's old results for a new domain.
 
-When no domain is chosen, offer to continue the connected company's saved work or evaluate a new public domain. Ask once for the research mission: the domain, intended buyer or market if known, and whether you may read its public pages and relevant public competitors. Within that scope, continue routine public reads and checks autonomously. Read [adaptive research](references/host-work.md#adaptive-research-and-tool-recovery) for priority-page understanding, tool discovery, interrupted inventories or a missing search/browser capability.
+When no domain is chosen, offer to continue the connected company's saved work or evaluate a new public domain. Ask once for the research mission: the domain, intended buyer or market if known, and whether you may read its public pages and relevant public competitors. Before any work is spent, run the [fit check](references/fit-and-scope.md): say plainly when the market is not measured or Obi is a weak fit, instead of proceeding. When the customer asks for several audiences, markets or languages, give the [one-scope answer](references/fit-and-scope.md#several-audiences-markets-or-languages). Within that scope, continue routine public reads and checks autonomously. Read [adaptive research](references/host-work.md#adaptive-research-and-tool-recovery) for priority-page understanding, tool discovery, interrupted inventories or a missing search/browser capability.
 
 Read `obi harness coverage --json` for the selected case before new Business work. If the account is inactive and `eligibleForFreeActivation` is true, the customer's agreed public-research mission includes activating the new Free plan with `obi harness activate --free --json`; show the actual question and engine coverage returned by that plan. An existing active plan needs no activation. If eligibility is false, keep its historical coverage and explain that an offer migration is separate. Activation does not start a scan. If its response is lost, read `harness coverage` before another activation attempt. When the customer wants to buy or change packs or engines, or evidence shows a gap the plan does not measure, read [purchase](references/purchase.md).
 
@@ -51,6 +51,8 @@ When the customer reports an implemented action, read [outcome observations](ref
 
 Lead with the outcome, material uncertainty and one useful next action. On an interrupted job, reread the same case and stage; report meaningful progress or a blocker, use the server's next poll time when available, and retain the work ID. A timeout on a write calls for a read before any retry. Reuse the same replay key for an identical result; a changed result needs the current work version and a new key.
 
+When a failure cannot be recovered in this conversation (the same error after the retry its `nextAction` names, or `obi doctor --json` reporting `server` as `degraded` or `unreachable`), say so and offer once to send it to Obi support. On the customer's yes, run `obi support "<what happened, in their words>" --json`: it adds the CLI version, the system and the last error code itself. When it fails with `SUPPORT_NOT_SENT`, nothing was sent: give them support@hireobi.com and the `support.diagnosticText` to paste. Never put a token or a credential file in the message.
+
 Separate site observations, customer confirmations, host inferences and independently measured answers. Request only the private extracts needed for the agreed scope; keep secrets and unrelated workspace material local. A local draft or patch is a reviewable suggestion until the customer accepts it. The Skill runs when invoked; while the harness is closed, saved server measurements can continue but host research and reasoning wait for resumption.
 
 Stay in customer mode even inside the Obi source repository. A product-use failure is explained and recovered in this conversation. Open a development task, branch or pull request only after an explicit separate software-development request.

package/skills/obi/references/fit-and-scope.md

@@ -0,0 +1,32 @@
+# Fit, markets and scope
+
+Read before the first research of a business, and when the customer asks for several audiences, markets or languages. Obi says what it cannot measure before any work is spent, and never stretches one scope to cover several.
+
+## Fit check, before any work
+
+Judge the fit from the domain, the customer's words and one read of the home page, before activating a plan or starting research. `obi capabilities --json` lists the markets Obi measures in `measurementMarkets`; name them from that list.
+
+| What you see | What you say | What you do next |
+|---|---|---|
+| The buyers are in a market outside `measurementMarkets` | "Obi does not measure <market> yet. It measures in: <the names from `measurementMarkets`>." | Ask whether they also sell in a listed market. If yes, offer that market as the scope and say the answers will describe that market only. If no, stop there. Either way, offer to record the request: `obi support "<market>, <domain>" --tag market-request --json`. |
+| The buyers search in a language no listed market uses | The same sentence, naming the language and the markets. | The same two offers. |
+| No AI-search buying journey: buyers choose without asking an assistant (an impulse or habit purchase, a walk-in local trade, a sale won only by tender or by marketplace listing) | "Obi is a weak fit for <business>: <the reason from what you read>. It measures what AI assistants answer to buyers who research before choosing." | Say what would change the verdict (a product line buyers research, a business customer segment). Continue only when the customer names such a segment or still asks for the Free measurement; then say what the scan will not show. |
+| The company has no public pages to read, or blocks them | "Obi reads public pages; it cannot ground questions without them." | Ask for the public domain that carries the offer; without one, stop. |
+
+`obi support` sends only from a signed-in machine. When it fails with `SUPPORT_NOT_SENT`, as for a prospect who has no account yet, give them support@hireobi.com and the `support.diagnosticText` to paste.
+
+A weak fit is a judgment you state with its evidence, as [business judgment](business-judgment.md) describes; the customer may correct the fact it rests on. Say it once, in one or two sentences, then the one next action. A supported market with a clear buying journey needs no fit statement.
+
+When the server answers `MARKET_UNSUPPORTED` to a questions submission, the same rule applies: say it plainly, name the markets, offer the request.
+
+## Several audiences, markets or languages
+
+One scope is one market, one primary audience and one measured language: the language buyers of that market use. An account holds one scope; packs add questions inside it. There is no second scope on an account today, so it has no price.
+
+When the customer asks for more (for example three ICPs, or three languages), answer in this order:
+
+1. The rule, in one sentence: "Obi measures one scope per account: one market, one primary audience, in that market's language. More packs add questions for that scope; they do not add a second one."
+2. Your recommendation of the primary scope, with its reason from the evidence: the segment the pricing page, customer stories and hiring serve, or the market where the measured rivals compete. Recommend one; ask only for confirmation.
+3. What still covers the others inside one scope: a secondary segment stays in the business review, and the three to five buyer profiles of the [cold and context pairs](questions.md#buyer-profiles-and-the-coldcontext-pairs) vary the personas of the primary audience.
+4. The limit, plainly: "A second market, audience or language on the same account is not available yet." Offer to record the request: `obi support "<the scopes wanted>" --tag scope-request --json`.
+5. Save the confirmed choice: `obi memory add --kind decision --text "<their words>" --json`.

package/skills/obi/references/purchase.md

@@ -1,6 +1,6 @@
 # Buying paid packs
 
-Read when the customer wants to buy packs or change the packs or engines of a paid plan, and before proposing a change yourself. Buying happens in this conversation, where the customer reviews the questions each pack will measure; the `/billing` page shows the plan and keeps cards, invoices and cancellation. Every price you show comes from the server's quote response.
+Read when the customer wants to buy packs or change the packs or engines of a paid plan, and before proposing a change yourself. Buying happens in this conversation, where the customer reviews the questions each pack will measure; the `/billing` page shows the plan and keeps cards, invoices and cancellation. Every amount the customer agrees to pay comes from the server's quote or preview response.
 
 Only the owner of the connected business account buys, with a verified email; a public-company evaluation case cannot.
 
@@ -70,7 +70,38 @@ A change that adds and removes at once (`PLAN_CHANGE_TWO_STEPS`) runs in two ste
 
 ## Proposing a change yourself
 
-When evidence shows a gap the plan does not measure, such as a buying situation outside the reviewed questions or a rival cited on questions the plan leaves out, propose the change once, with that evidence and the price from a quote or preview. After a no, raise it again only on new evidence.
+Propose a change only when a measurement shows a gap the plan does not cover. Each trigger needs its evidence from the saved results (`obi questions results --json`, `obi harness evidence`), named by question and date:
+
+| Trigger | Measured evidence it needs | What you propose |
+|---|---|---|
+| Perplexity | On research-style buying questions (comparisons, alternatives, "which should we use"), the scan's answers name rivals with cited sources while the company is absent or uncited. | Perplexity on those questions, 1 to 10 per pack. |
+| Claude | The confirmed primary audience is technical or B2B, and on the same measured buying questions the engines already disagree on the brands they name. Where Claude is already measured on some questions, its answers differ from the other engines on them. | Claude on those buying questions, 1 to 10 per pack. |
+| More packs | Buying situations in the evidence (a confirmed segment's situations, a rival's angle, the sources the answers cite) that no reviewed question covers, while every current question covers a distinct one. | One more pack, with the situations it would add. |
+| Paid from Free | Two monthly scans differ (a rival enters or leaves the answers, the company's presence changes) and nothing dates the change inside the month, or the customer reported an action whose effect the next monthly scan would show weeks late. | The first pack, for its weekly scan. |
+
+The host's own answer to a question is never this evidence, and an engine the plan does not measure has no measured answers to cite.
+
+A proposal is one sentence of evidence, the price, one question. Cite the server's price whenever a quote is possible; both commands below charge nothing:
+
+| Plan and change | Price to cite |
+|---|---|
+| A paid plan, adding an engine on questions it already measures | Write the quote file of the new plan and run `obi harness purchase change --input <file> --json` ([changing a paid plan](#changing-a-paid-plan), step 2): cite `amountDueCents` now and the new `quote.monthlyAmountCents`. |
+| No paid plan, with a reviewed set of 25 questions per pack | `obi harness purchase quote --input <file> --json`: cite `monthlyAmountCents`. |
+| The change needs questions not reviewed yet (one more pack, a first pack without a reviewed set) | No quote is possible yet: cite the list price below, labelled "list price", and offer to prepare the questions and the quote. |
+
+The list price is what the change adds each month, computed from `obi capabilities --json` (`catalog.monthlyCents`) and the plan in `obi harness coverage --json`:
+
+| Change | Monthly amount it adds |
+|---|---|
+| An engine on a paid plan | Its per-pack amount times the packs of the plan (`plan.questionCount` divided by 25). |
+| One more pack | `corePack`, plus the per-pack amount of each engine already on the plan (an engine whose `plan.engineQuestionCounts` is above zero). |
+| The first pack from Free | `corePack`. |
+
+A list price is before any discount on the account; the quote or preview confirms the exact amount before any yes to pay. For example: "On 6 of your 10 comparison questions, the 3 November scan cites <rival> with sources and never you. Perplexity on those questions adds <monthlyAmountCents difference> a month, <amountDueCents> today. Shall I apply it?"
+
+- Propose once per gap, and at most one proposal in a reply. After a no, save it (`obi memory add --kind decision --text "<their words>" --json`) and raise that gap again only on new evidence: a later scan, a reported outcome, a corrected business fact.
+- Optional engines are never filled to quota: propose the buying questions the evidence names, however few.
+- No proposal while a failure is unresolved, before the first results exist, or as the answer to a question the customer asked about something else.
 
 ## What waits for the customer
 

package/skills/obi/references/questions.md

@@ -1,6 +1,6 @@
 # Buyer questions, scan and priorities
 
-After the business review is confirmed, inspect `obi harness show --stage questions --json`. Show the supported measurement markets and confirmed buyer audiences from its work context. Ask the customer to choose one market and one primary audience; a suggested market is guidance, not consent or a default. Prepare the displayed coverage count in the host and save it with the versioned `harness submit --stage questions --input <file> --json` path described in [host work](host-work.md). `obi questions show --json` reads the accepted set and its preparation state. A queued or running state means wait and inspect again after meaningful progress; do not announce that questions already exist.
+After the business review is confirmed, inspect `obi harness show --stage questions --json`. Show the supported measurement markets and confirmed buyer audiences from its work context. Ask the customer to choose one market and one primary audience; a suggested market is guidance, not consent or a default. When they ask for several, give the [one-scope answer](fit-and-scope.md#several-audiences-markets-or-languages). Prepare the displayed coverage count in the host and save it with the versioned `harness submit --stage questions --input <file> --json` path described in [host work](host-work.md). `obi questions show --json` reads the accepted set and its preparation state. A queued or running state means wait and inspect again after meaningful progress; do not announce that questions already exist.
 
 ## Writing the questions