@marginfront/code-cost-clarity 0.5.4 → 0.6.0

package/README.md

@@ -2,7 +2,7 @@
 
 > See your Claude Code **and Codex** spend in MarginFront. One command wires your
 > coding-agent usage telemetry through a local collector and into MarginFront,
-> priced per engineer, per model, with **accurate prompt-cache token splitting**.
+> priced per developer, per model, with **accurate prompt-cache token splitting**.
 > One package, three modes: Claude Code only, Codex only, or both.
 
 > **⚠️ Beta (pilot software).** Under active development - pin a version, expect rough
@@ -17,8 +17,8 @@
 **📖 Full documentation:** https://docs.marginfront.com/tools/code-cost-clarity
 
 **This is internal cost visibility, not billing and not a spend cap.** The
-installing company watches its own AI coding spend (per engineer, per model,
-R&D-vs-COGS). It does not charge engineers and it does not cut anyone off.
+installing company watches its own AI coding spend (per developer, per model,
+R&D-vs-COGS). It does not charge developers and it does not cut anyone off.
 
 **Your LLM keys stay yours.** This tool never reads, needs, or transmits your
 Anthropic or OpenAI API key. The only credential it uses is your MarginFront key
@@ -28,7 +28,7 @@ Anthropic or OpenAI API key. The only credential it uses is your MarginFront key
 
 ## What this does (one sentence)
 
-Every time an engineer runs Claude Code, this connector captures the token usage
+Every time a developer runs Claude Code, this connector captures the token usage
 and sends it to MarginFront as a usage event, so you can see who used what, on
 which model, and how much it cost.
 
@@ -37,13 +37,13 @@ which model, and how much it cost.
 Think of it like a cash-register receipt system:
 
 1. **Claude Code** is the register. As it works, it broadcasts receipts
-   (OpenTelemetry exports): how many tokens, which model, which engineer.
+   (OpenTelemetry exports): how many tokens, which model, which developer.
 2. **The collector** (`otelcol-contrib`, open source) is the catcher. It runs in
    the background, catches the receipts, and writes them to a file.
 3. **The forwarder** (our glue, inside this package) reads those receipts and
    sends each one to MarginFront.
 4. **MarginFront** records the event, prices it, and shows it under the
-   engineer's email.
+   developer's email.
 
 ```
 Claude Code (your task)
@@ -103,7 +103,7 @@ claude          # or: codex   (or open the Claude/Codex desktop apps)
 > config when they launch, so anything that was already open won't emit until it's reopened.
 
 That's it. No `source`, no second terminal. Your spend shows up in MarginFront under
-your engineer email, automatically, and the background meter restarts itself at
+your developer email, automatically, and the background meter restarts itself at
 every login. Check on it anytime:
 
 ```bash
@@ -115,7 +115,7 @@ npx @marginfront/code-cost-clarity status
 stops it). You'll see a line per turn like:
 
 ```
-[14:22:07] recorded engineer@example.com · in=3210 out=287 · server=$0.0232 · cc=$0.0236 event=9f0c2a71-...
+[14:22:07] recorded developer@example.com · in=3210 out=287 · server=$0.0232 · cc=$0.0236 event=9f0c2a71-...
 ```
 
 **CI / scripted installs:** add `--no-prompt` to `init` to skip BOTH questions (the
@@ -180,34 +180,75 @@ It does **not** capture:
 
 ---
 
-## Per-engineer attribution is automatic
+## Per-developer attribution is automatic
 
 What makes "who spent what" work is `user.email`, and **Claude Code puts it in
-the telemetry on its own**. It's the engineer's logged-in Claude account email.
-There is no manual email config. Each engineer loads the telemetry settings
+the telemetry on its own**. It's the developer's logged-in Claude account email.
+There is no manual email config. Each developer loads the telemetry settings
 and runs Claude Code normally.
 
-**Works whether the engineer signs in with an org-managed seat or an interactive
+**Works whether the developer signs in with an org-managed seat or an interactive
 login.** On org-managed Claude seats the email is stamped for free. If an
-engineer's sign-in doesn't surface an email, the connector doesn't drop the
+developer's sign-in doesn't surface an email, the connector doesn't drop the
 usage. It attributes it to a clearly labeled placeholder customer
 (`claude-code-no-identity`) and prints how to fix it (sign in with an org-managed
 seat, or attach a customer mapping). You'll see the placeholder in `preview`/`run`
 output if it ever kicks in.
 
-> The MarginFront **API key** is separate from the engineer's identity: it's the
+> The MarginFront **API key** is separate from the developer's identity: it's the
 > connector's own credential for posting to MarginFront. `run` needs it;
 > `preview` does not.
 
 ---
 
+## Shared one login across the team? Set a per-machine developer (`CCC_DEVELOPER_EMAIL`)
+
+Some teams share **one** Claude/Codex login across the whole team. When that
+happens, every developer's telemetry carries the **same** `user.email`, so all the
+AI cost collapses onto one person and per-developer attribution breaks. CCC fixes
+this without any change to how you log in, because **CCC runs locally on each
+laptop** - so each laptop can carry its own developer identity.
+
+**During `init` we ask one question:** which developer this machine's AI cost should
+be attributed to. The prompt is pre-filled with your **global git email**
+(`git config --global user.email`), which is usually the right person sitting at
+this laptop. You have three choices:
+
+- **Press Enter** to accept the pre-filled email.
+- **Type a different email** to attribute this machine to someone else.
+- **Clear it and leave it blank** to keep the old behavior - auto-detect from
+  whatever email the coding-agent login reports.
+
+Whatever you choose is saved as **`CCC_DEVELOPER_EMAIL`** in
+`~/.marginfront-ccc/.env` (mode 600, right next to your key). To change it later,
+edit that one line and then `stop` + `start` (or just re-run `init`).
+
+**Precedence at send time (highest wins):**
+
+1. **`CCC_DEVELOPER_EMAIL`** (this machine's developer) - if set, it wins for **every**
+   record this machine sends, Claude **and** Codex, token turns **and** tool calls.
+2. Otherwise the email the coding-agent login reports (`user.email`).
+3. Otherwise the no-identity placeholder (`claude-code-no-identity` /
+   `codex-no-identity`).
+
+If you **don't** set it (leave it blank), nothing changes - attribution is
+byte-for-byte what it was before this feature: the login's own `user.email`, then
+the placeholder.
+
+> This is **internal cost visibility only**. `CCC_DEVELOPER_EMAIL` re-labels which
+> developer the cost shows up under in _your_ MarginFront - it does **not** change
+> who pays or any customer's bill. The `--no-prompt` (CI / scripted) install skips
+> this question and leaves `CCC_DEVELOPER_EMAIL` unset.
+
+---
+
 ## Privacy (read this)
 
 This tool is **internal cost visibility**, and being honest about what it turns on
 matters, so here is the full picture in plain English.
 
-**Your engineer email travels in the telemetry, in plaintext, on purpose.** That's
-the whole point: per-engineer cost attribution needs to know who ran the turn.
+**Your developer email travels in the telemetry, in plaintext, on purpose.** That's
+the whole point: per-developer cost attribution needs to know who ran the turn.
 Claude Code stamps `user.email` (your logged-in account email) onto the usage
 telemetry, the local collector writes it to a file on your machine, and the
 forwarder POSTs it to MarginFront so the spend lands under your name. It is **not**
@@ -304,7 +345,7 @@ silent charge). Free built-ins (file reads, shell, grep) are never forwarded.
 
 The same package captures **Codex** too. Nothing extra to install. Codex reports
 to the **same** local collector. You get three modes for free: Claude Code only,
-Codex only, or **both** (same engineer's name on all of it).
+Codex only, or **both** (same developer's name on all of it).
 
 **`init` turns it on for you** when you consent, it adds this `[otel]` block to
 your `~/.codex/config.toml` automatically:
@@ -350,8 +391,8 @@ the input rate and drops the cache-read field, so the number reads high, never l
 
 ### Identity by sign-in mode (org seat vs ChatGPT login)
 
-Per-engineer attribution rides on `user.email` from Codex's telemetry, same idea
-as Claude Code. Whether that email surfaces depends on how the engineer signs in
+Per-developer attribution rides on `user.email` from Codex's telemetry, same idea
+as Claude Code. Whether that email surfaces depends on how the developer signs in
 to Codex (an org-managed or API-key sign-in stamps it; some interactive ChatGPT
 logins may not). We never read that OpenAI key; the only thing that matters here is
 whether the telemetry carries an email. If it doesn't surface, the usage isn't
@@ -375,7 +416,7 @@ exactly like the Claude path.
 
 > **Confirm the dollar figure once.** The exact Codex dollar figure should be
 > confirmed against one real captured session: that Codex reports per-turn (not
-> session-cumulative) token counts on the log stream, and that your engineer email
+> session-cumulative) token counts on the log stream, and that your developer email
 > populates under your sign-in mode. The mapping above is the verified-safe
 > default; the confirmation is a one-session spike, not a blocker to installing.
 
@@ -454,7 +495,7 @@ curl -s -H "x-api-key: $KEY" \
 
 ---
 
-## For engineers (technical appendix)
+## For developers (technical appendix)
 
 **Input shape:** OTLP/JSON, tree `resourceMetrics[].scopeMetrics[].metrics[]`. Two
 metrics matter: `claude_code.token.usage` (one datapoint per `type` in

package/dist/cli.d.ts

@@ -33,6 +33,7 @@ interface PromptIO {
     terminal?: boolean;
 }
 declare function promptYesNo(promptText: string, io?: PromptIO, defaultYes?: boolean): Promise<boolean>;
+declare function promptVisible(promptText: string, defaultValue?: string, io?: PromptIO): Promise<string>;
 interface InitDeps {
     ensureConfigFiles?: () => void;
     ensureCollectorBinary?: () => void;
@@ -40,6 +41,9 @@ interface InitDeps {
     isTTY?: () => boolean;
     promptSecret?: (promptText: string) => Promise<string>;
     writeApiKey?: (value: string) => void;
+    gitDefaultEmail?: () => string;
+    promptDeveloperEmail?: (defaultValue: string) => Promise<string>;
+    writeDeveloperEmail?: (value: string) => void;
     hasApiKey?: () => boolean;
     promptConsent?: () => Promise<boolean>;
     writeClaudeTelemetry?: () => WriteClaudeSettingsResult;
@@ -87,4 +91,4 @@ interface UninstallDeps {
 }
 declare function cmdUninstall(purge: boolean, deps?: UninstallDeps): number;
 
-export { type InitDeps, type PromptIO, type StartDeps, type StopDeps, type UninstallDeps, cmdInit, cmdRun, cmdStart, cmdUninstall, promptYesNo, runningForwarderPid, stopForwarderThenStopCollector };
+export { type InitDeps, type PromptIO, type StartDeps, type StopDeps, type UninstallDeps, cmdInit, cmdRun, cmdStart, cmdUninstall, promptVisible, promptYesNo, runningForwarderPid, stopForwarderThenStopCollector };

package/dist/cli.js

@@ -102,6 +102,21 @@ OTEL_METRIC_EXPORT_INTERVAL=300000
 # MarginFront's pricing catalog decides what's billable and at what rate. Add price
 # rows for your paid tools in MarginFront - there's no client-side list to maintain.
 
+# WHO this machine's AI cost belongs to (per-developer attribution).
+# WHY THIS EXISTS, plainly: lots of teams share ONE Claude/Codex login, so every
+# developer's usage carries the SAME login email and all the cost piles onto one
+# person. CCC runs locally on each laptop, so naming the developer here splits that
+# shared-login cost back out per developer. This is INTERNAL cost visibility only -
+# it does NOT change anyone's bill.
+#
+# HOW IT'S USED at send time (precedence, highest wins):
+#   1. This CCC_DEVELOPER_EMAIL, when set - it wins for EVERY record from this machine.
+#   2. Otherwise the email the coding-agent login reports (user.email).
+#   3. Otherwise a clearly-labeled no-identity placeholder.
+# Leave it BLANK to keep today's behavior (auto-detect from the login). init fills
+# this in for you, defaulting to your global git user.email - change it anytime.
+CCC_DEVELOPER_EMAIL=
+
 # YOUR MarginFront SECRET key (mf_sk_*). Create or copy one at:
 #   app.marginfront.com -> Build -> API Keys -> "Create Key Pair"
 #   (https://app.marginfront.com/developer-zone/api-keys)
@@ -205,7 +220,7 @@ function renderCodexConfigInstructions() {
     "  Codex, or both - from the one collector file; there's no extra flag.",
     "",
     "  Note: exact [otel] key names can vary by Codex version, and whether your",
-    "  engineer email shows up depends on how you sign in (org API key vs ChatGPT",
+    "  developer email shows up depends on how you sign in (org API key vs ChatGPT",
     "  login). If the email doesn't surface, usage still lands under the no-identity",
     "  placeholder. See the README's Codex section for the full details."
   ].join("\n");
@@ -447,6 +462,32 @@ function writeApiKeyToEnv(value, path = ENV_PATH) {
   writeFileSync(path, next, { mode: 384 });
   chmodSync(path, 384);
 }
+function gitGlobalUserEmail() {
+  try {
+    const out = execFileSync("git", ["config", "--global", "user.email"], {
+      encoding: "utf8",
+      // Swallow git's own stderr so a "key not set" message can't leak to the user.
+      stdio: ["ignore", "pipe", "ignore"]
+    });
+    return out.trim();
+  } catch {
+    return "";
+  }
+}
+function writeDeveloperEmailToEnv(value, path = ENV_PATH) {
+  const current = existsSync(path) ? readFileSync(path, "utf8") : renderEnvFile();
+  const developerLine = /^CCC_DEVELOPER_EMAIL=.*$/m;
+  let next;
+  if (developerLine.test(current)) {
+    next = current.replace(developerLine, () => `CCC_DEVELOPER_EMAIL=${value}`);
+  } else {
+    const sep = current.endsWith("\n") || current === "" ? "" : "\n";
+    next = `${current}${sep}CCC_DEVELOPER_EMAIL=${value}
+`;
+  }
+  writeFileSync(path, next, { mode: 384 });
+  chmodSync(path, 384);
+}
 var CLAUDE_SETTINGS_PATH = join(homedir(), ".claude", "settings.json");
 var CLAUDE_SETTINGS_OWNERSHIP_PATH = join(
   CONFIG_DIR,
@@ -952,8 +993,8 @@ import { createRequire } from "module";
 import process2 from "process";
 var _cccRequire = createRequire(import.meta.url);
 function resolveCccPackageVersion() {
-  if ("0.5.4".length > 0) {
-    return "0.5.4";
+  if ("0.6.0".length > 0) {
+    return "0.6.0";
   }
   try {
     const pkg = _cccRequire("../package.json");
@@ -981,6 +1022,9 @@ function resolveIngestUrl(override = process2.env.MARGINFRONT_INGEST_URL) {
   return DEFAULT_INGEST_URL;
 }
 var MARGINFRONT_INGEST_URL = resolveIngestUrl();
+function resolveDeveloperOverride(env) {
+  return (env.CCC_DEVELOPER_EMAIL || "").trim() || void 0;
+}
 var AGENT_CODE = "claude-code";
 var SIGNAL_NAME = "claude-code-turn";
 var MODEL_PROVIDER = "anthropic";
@@ -1097,6 +1141,7 @@ function idempotencyKeyFor(rawSourceLine, email, rawModel, sessionId, indexWithi
 function buildMarginFrontRequestBody(parsedOtlp, options = {}) {
   const foldCache = options.foldCache === true;
   const fallbackCustomerId = typeof options.fallbackCustomerId === "string" && options.fallbackCustomerId.length > 0 ? options.fallbackCustomerId : null;
+  const developerOverride = options.developerOverride;
   const tokenDataPoints = collectDataPointsForMetric(
     parsedOtlp,
     "claude_code.token.usage"
@@ -1113,7 +1158,7 @@ ${sessionId}`;
   }
   for (const dataPoint of tokenDataPoints) {
     const attributes = attributesToLookup(dataPoint.attributes);
-    const email = attributes["user.email"] || fallbackCustomerId;
+    const email = developerOverride || attributes["user.email"] || fallbackCustomerId;
     const rawModel = attributes["model"];
     const sessionId = attributes["session.id"];
     const tokenType = attributes["type"];
@@ -1146,7 +1191,7 @@ ${sessionId}`;
   function findCostForGroup(email, rawModel, sessionId) {
     for (const costPoint of costDataPoints) {
       const costAttributes = attributesToLookup(costPoint.attributes);
-      const costEmail = costAttributes["user.email"] || fallbackCustomerId;
+      const costEmail = developerOverride || costAttributes["user.email"] || fallbackCustomerId;
       if (costEmail === email && costAttributes["model"] === rawModel && costAttributes["session.id"] === sessionId) {
         return typeof costPoint.asDouble === "number" ? costPoint.asDouble : null;
       }
@@ -1167,7 +1212,7 @@ ${sessionId}`;
       group.sessionId
     );
     const record = {
-      // Per-engineer attribution: the engineer's email IS the customer id (or the
+      // Per-developer attribution: the developer's email IS the customer id (or the
       // no-identity placeholder).
       customerExternalId: group.email,
       agentCode: AGENT_CODE,
@@ -1245,6 +1290,7 @@ function isCodexUsageRecord(lookup, eventName) {
 function buildCodexRequestBody(parsedOtlp, options = {}) {
   const foldCache = options.foldCache === true;
   const fallbackCustomerId = typeof options.fallbackCustomerId === "string" && options.fallbackCustomerId.length > 0 ? options.fallbackCustomerId : CODEX_NO_IDENTITY_CUSTOMER;
+  const developerOverride = options.developerOverride;
   const records = [];
   const resourceLogsList = parsedOtlp?.resourceLogs;
   if (!Array.isArray(resourceLogsList)) return { records };
@@ -1262,7 +1308,7 @@ function buildCodexRequestBody(parsedOtlp, options = {}) {
         const rawModel = logAttrString(lookup["model"]);
         if (!rawModel) continue;
         if (CODEX_EXCLUDED_MODELS.has(rawModel.trim().toLowerCase())) continue;
-        const email = logAttrString(lookup["user.email"]) || fallbackCustomerId;
+        const email = developerOverride || logAttrString(lookup["user.email"]) || fallbackCustomerId;
         if (!email) continue;
         const sessionId = logAttrString(lookup["session.id"]) ?? logAttrString(lookup["conversation.id"]) ?? logAttrString(lookup["session_id"]) ?? null;
         const inputCount = logAttrTokenCount(lookup[CODEX_INPUT_TOKEN_KEY]);
@@ -1327,6 +1373,7 @@ function toolResultSource(eventName) {
 }
 function buildToolUsageRecords(parsedOtlp, options = {}) {
   const fallbackCustomerId = typeof options.fallbackCustomerId === "string" && options.fallbackCustomerId.length > 0 ? options.fallbackCustomerId : CODEX_NO_IDENTITY_CUSTOMER;
+  const developerOverride = options.developerOverride;
   const records = [];
   const resourceLogsList = parsedOtlp?.resourceLogs;
   if (!Array.isArray(resourceLogsList)) return { records };
@@ -1352,7 +1399,7 @@ function buildToolUsageRecords(parsedOtlp, options = {}) {
         }
         if (!toolName) continue;
         if (BUILTIN_NONBILLABLE_TOOLS.has(toolName)) continue;
-        const email = logAttrString(lookup["user.email"]) || fallbackCustomerId;
+        const email = developerOverride || logAttrString(lookup["user.email"]) || fallbackCustomerId;
         if (!email) continue;
         const sessionId = logAttrString(lookup["session.id"]) ?? logAttrString(lookup["conversation.id"]) ?? logAttrString(lookup["session_id"]) ?? null;
         const groupKey = `${email}
@@ -1618,6 +1665,7 @@ function watchAndForward(filePath, options = {}) {
   const fallbackCustomerId = options.fallbackCustomerId;
   const cursorPath = options.cursorPath;
   const pidFilePath = options.pidFilePath;
+  const developerOverride = resolveDeveloperOverride(process2.env);
   if (!process2.env.MARGINFRONT_API_KEY) {
     console.error(
       "MARGINFRONT_API_KEY is not set. Set it in your shell before watching: export MARGINFRONT_API_KEY=..."
@@ -1751,6 +1799,7 @@ function watchAndForward(filePath, options = {}) {
         const body = buildRequestBodyForLine(parsed, {
           foldCache,
           fallbackCustomerId,
+          developerOverride,
           rawSourceLine: line
         });
         if (!body.records.length) continue;
@@ -1820,8 +1869,10 @@ function printHelp() {
       "",
       "Commands:",
       "  init                 Set up everything: save your MarginFront secret key (mf_sk_...), not",
-      "                       your Anthropic or OpenAI key. Wire telemetry into your Claude/Codex",
-      "                       config (with your OK), and start the background meter.",
+      "                       your Anthropic or OpenAI key. Ask which developer THIS machine's AI",
+      "                       cost belongs to (for teams on a shared Claude/Codex login). Wire",
+      "                       telemetry into your Claude/Codex config (with your OK), and start",
+      "                       the background meter.",
       "  start                (Re)start the background meter - starts at login, no terminal to keep open.",
       "  status               Is the background meter running? Shows recent activity + errors.",
       "  preview <file.json>  Show the exact record it would send for a capture. No key needed.",
@@ -1846,19 +1897,29 @@ function printHelp() {
       "  automatically. No `source`, no second terminal. Check on it with `status` anytime.",
       "",
       "Prefer a live terminal view instead of the background meter? After init, run:",
-      "  npx @marginfront/code-cost-clarity run"
+      "  npx @marginfront/code-cost-clarity run",
+      "",
+      "Sharing one Claude/Codex login across a team? (per-developer cost attribution)",
+      "  During `init` we ask which developer THIS machine's AI cost belongs to and save it",
+      "  as CCC_DEVELOPER_EMAIL in ~/.marginfront-ccc/.env. At send time the order is:",
+      "    1. CCC_DEVELOPER_EMAIL (this machine's developer) - if set, it always wins;",
+      "    2. otherwise the email the coding-agent login reports;",
+      "    3. otherwise a no-identity placeholder.",
+      "  Leave it BLANK to keep the old auto-detect behavior. Edit the .env to change it,",
+      "  then `stop` and `start` (or just re-run `init`). Internal cost visibility only -",
+      "  it does NOT change anyone's bill."
     ].join("\n")
   );
 }
 function printNoIdentityHint() {
   console.log(
     [
-      "Note: usage with no engineer email is attributed to a no-identity placeholder",
+      "Note: usage with no developer email is attributed to a no-identity placeholder",
       `  ("${NO_IDENTITY_CUSTOMER}" for Claude Code, "${CODEX_NO_IDENTITY_CUSTOMER}" for Codex).`,
       "  Why: org-managed seats stamp the email automatically; some interactive logins don't.",
       "  Fix: sign in with an org-managed seat, or attach your own MarginFront customer",
       "  mapping. Until then the spend still lands, under the placeholder instead of the",
-      "  engineer's name."
+      "  developer's name."
     ].join("\n")
   );
 }
@@ -1912,6 +1973,30 @@ function promptYesNo(promptText, io = {}, defaultYes = false) {
     rl.on("close", () => finish("", false));
   });
 }
+function promptVisible(promptText, defaultValue = "", io = {}) {
+  return new Promise((resolve) => {
+    const input = io.input ?? process3.stdin;
+    const output = io.output ?? process3.stdout;
+    const interactive = io.terminal ?? Boolean(input.isTTY);
+    const rl = readline.createInterface({
+      input,
+      output,
+      terminal: interactive
+    });
+    let settled = false;
+    const finish = (answer) => {
+      if (settled) return;
+      settled = true;
+      rl.close();
+      resolve(answer.trim());
+    };
+    output.write(promptText);
+    rl.question("", (answer) => finish(answer));
+    if (interactive && defaultValue) rl.write(defaultValue);
+    rl.on("SIGINT", () => finish(""));
+    rl.on("close", () => finish(""));
+  });
+}
 function printConsentDisclosure(log) {
   log(
     [
@@ -1987,6 +2072,41 @@ async function cmdInit(noPrompt, deps = {}) {
       keySaved = true;
     }
   }
+  const shouldPromptDeveloper = !noPrompt && isTTY();
+  if (shouldPromptDeveloper) {
+    const gitDefault = (deps.gitDefaultEmail ?? (() => gitGlobalUserEmail()))();
+    const savedDeveloperEmail = loadEnv()["CCC_DEVELOPER_EMAIL"];
+    const defaultEmail = savedDeveloperEmail || gitDefault;
+    log(
+      [
+        "",
+        "Who should this machine's AI cost be attributed to?",
+        "  If your team shares ONE Claude/Codex login, every developer's usage looks",
+        "  like the same person. Naming the developer here splits this laptop's spend",
+        "  back out to them. (Internal cost visibility only - it does NOT change anyone's bill.)",
+        "",
+        "  - Press Enter to accept the pre-filled email (from your global git config).",
+        "  - Type a different email to attribute this machine to someone else.",
+        "  - Clear it and leave it BLANK to auto-detect from your coding-agent login",
+        "    (today's behavior - usage is tagged with whatever email the login reports).",
+        ""
+      ].join("\n")
+    );
+    const askDeveloper = deps.promptDeveloperEmail ?? ((defaultValue) => promptVisible(
+      "Developer email for this machine's AI cost: ",
+      defaultValue
+    ));
+    const chosenEmail = (await askDeveloper(defaultEmail)).trim();
+    const writeDeveloper = deps.writeDeveloperEmail ?? ((value) => writeDeveloperEmailToEnv(value));
+    writeDeveloper(chosenEmail);
+    if (chosenEmail) {
+      log(`Will attribute this machine's AI cost to ${chosenEmail}.`);
+    } else {
+      log(
+        "Left attribution on auto-detect (this machine's AI cost follows the coding-agent login)."
+      );
+    }
+  }
   printConsentDisclosure(log);
   let consented;
   if (noPrompt) {
@@ -2490,6 +2610,7 @@ export {
   cmdRun,
   cmdStart,
   cmdUninstall,
+  promptVisible,
   promptYesNo,
   runningForwarderPid,
   stopForwarderThenStopCollector

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@marginfront/code-cost-clarity",
-  "version": "0.5.4",
-  "description": "See your Claude Code and Codex spend in MarginFront. One command wires your coding-agent usage telemetry through a local collector into MarginFront, priced per engineer, per model, with accurate prompt-cache token splitting and automatic billable tool-call tracking.",
+  "version": "0.6.0",
+  "description": "See your Claude Code and Codex spend in MarginFront. One command wires your coding-agent usage telemetry through a local collector into MarginFront, priced per developer, per model, with accurate prompt-cache token splitting and automatic billable tool-call tracking.",
   "type": "module",
   "bin": {
     "code-cost-clarity": "dist/cli.js"