fix(evals): support explicit Grok qualification workflows (#13878)

## Thinking Path > - Paperclip manages AI agents and their provider connections. > - Product E2E checks real tasks through the browser, server, and runner. > - Grok qualification needs separate API-key and subscription evidence. > - Product subscription tests and direct Grok protocol evals need explicit credential delivery. > - This change supplies each credential only to its selected profile and prepares the pinned binary. > - Maintainer authorization and protected-environment gates remain required. ## Linked Issues or Issue Description Refs #13845, #13847, #13850, #13882. The Grok feature branch has a manual subscription qualification profile. The trusted master workflow must admit its selected credential and prepare the same verified binary and artifact verifier as the API profile. Direct protocol evals also need the selected xAI key and pinned Grok binary. These prerequisites do not register or schedule the new profiles on master. ## What Changed - Deliver `GROK_AUTH_JSON` from the protected paid environment only when the selected profile requests that credential. - Install the checksum-verified Grok binary for the local subscription profile. - Prepare the pinned artifact verifier for the manual subscription suite. - Extend workflow security assertions to cover the new credential and profile. - Add the ACPX Grok credential mapping to the trusted-master catalog, then deliver only the selected `XAI_API_KEY` to direct protocol cells and install the target’s checksum-verified Grok binary before packaging. - Allow a direct-protocol concurrency override from two cases up to the existing configured ceiling; it can only lower concurrency. - Document the Grok protocol workflow and its API-only credential boundary. - Render missing LLM usage and cost as Unavailable, and label partial observations with coverage. Preserve raw records, grades, and actual zero costs. - Preserve measured campaign source metadata during report regeneration instead of inheriting the renderer checkout or CI event; skip empty legacy source records when recovering older provenance. ## Verification - Latest commit `05d05801477104c8155977bbbe3e119a5241f960`: all 54 reported checks successful, two intentional skips, Greptile 5/5, and zero unresolved review threads. [CI run](https://github.com/paperclipai/paperclip/actions/runs/35890978288). - After merging current master, all 17 workflow security/image tests and 23 catalog/workflow policy tests passed. The trusted catalog also generated all 39 pinned Grok cells with `XAI_API_KEY` and one case per shard. The new policy tests execute the concurrency guard against valid, out-of-range, and malformed values. - The Grok branch separately passed 450 Product harness unit tests, including private company credential staging, cleanup, and token-fragment redaction. - The fresh-login native subscription smoke passed three repetitions of tool execution, session resume, restrictive permissions, and cleanup. These are setup evidence; full subscription Product qualification remains pending. - All 72 focused report/billing/history/catalog tests and the Product harness typecheck passed for the report-display change. The initial sandbox run could not open the tsx IPC socket; the permitted rerun passed. A zero-provider-call replay of the actual 16-cell Grok campaign preserved all result records, grades, timing, and source provenance while correcting missing usage labels. - Review the thirteen-file diff. Provider credentials still enter only the selected paid-test step; default-branch, numeric-actor, and environment restrictions are unchanged. ## Risks This admits a refreshable subscription credential to explicitly selected trusted tests. Store it only in `runner-e2e-paid`, use a test login, and remove it after qualification. Unselected profiles receive an empty value. Pull requests cannot trigger the paid workflow. This PR changes no fleet admission or actor allowlist. ## Model Used OpenAI GPT-6 through Codex, with tool use and code execution. The exact serving model identifier and context-window size are not exposed in this session. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-10-06 10:48:12 +02:00 · 2026-09-23 17:05:52 -05:00
1 parent 4b8ec588f3
commit b648d8cdda
@@ -891,7 +891,7 @@ jobs:
        run: node packages/paperclip-runner/scripts/materialize-opencode-binary.mjs

      - name: Install checksum-verified Grok executable
-        if: matrix.environmentId == 'local' && matrix.profileId == 'runner-acpx-grok'
+        if: matrix.environmentId == 'local' && (matrix.profileId == 'runner-acpx-grok' || matrix.profileId == 'runner-acpx-grok-subscription')
        run: node packages/grok-acp/install.mjs

      - name: Download immutable campaign outputs
@@ -1044,7 +1044,7 @@ jobs:
          NODE

      - name: Prepare pinned Python artifact oracle image
-        if: (matrix.suiteId == 'everyday-workflows' || matrix.suiteId == 'grok-qualification') && (matrix.caseId == 'build-revise' || matrix.caseId == 'delegate-feedback' || matrix.caseId == 'agent-review-handoff' || matrix.caseId == 'hire-reuse' || matrix.caseId == 'recover-controller' || matrix.caseId == 'stop-redirect')
+        if: (matrix.suiteId == 'everyday-workflows' || matrix.suiteId == 'grok-qualification' || matrix.suiteId == 'grok-subscription-qualification') && (matrix.caseId == 'build-revise' || matrix.caseId == 'delegate-feedback' || matrix.caseId == 'agent-review-handoff' || matrix.caseId == 'hire-reuse' || matrix.caseId == 'recover-controller' || matrix.caseId == 'stop-redirect')
        run: |
          set -euo pipefail
          oracle_image='python@sha256:9d2e5553305c7c7b0097999bb17187c69b921ccd6bc9d40e4bb5ebe652c00285'
@@ -1058,6 +1058,7 @@ jobs:
          ANTHROPIC_API_KEY: ${{ matrix.credentialName == 'ANTHROPIC_API_KEY' && secrets.ANTHROPIC_API_KEY || '' }}
          OPENROUTER_API_KEY: ${{ matrix.credentialName == 'OPENROUTER_API_KEY' && secrets.OPENROUTER_API_KEY || '' }}
          XAI_API_KEY: ${{ matrix.credentialName == 'XAI_API_KEY' && secrets.XAI_API_KEY || '' }}
+          GROK_AUTH_JSON: ${{ matrix.credentialName == 'GROK_AUTH_JSON' && secrets.GROK_AUTH_JSON || '' }}
          DAYTONA_API_KEY: ${{ matrix.environmentId == 'daytona' && secrets.DAYTONA_API_KEY || '' }}
          PAPERCLIP_E2E_DAYTONA_IMAGE: ${{ needs.daytona_image.outputs.image }}
          PAPERCLIP_RUNNER_REMOTE_PROVIDER_PACK_PATH: ${{ github.workspace }}/packages/paperclip-runner/provider-pack
@@ -18,6 +18,10 @@ on:
        type: string
        default: "all"
        required: false
+      max_parallel:
+        description: "Optional lower concurrency (at least 2, no higher than the configured campaign limit)"
+        type: string
+        required: false
      max_infrastructure_retries:
        description: "Automatic retries only for explicitly retryable infrastructure failures (0-3)"
        type: number
@@ -250,6 +254,7 @@ jobs:
          PAPERCLIP_PROTOCOL_EVALS_SHA: ${{ needs.authorize.outputs.evals_sha }}
          MAX_PARALLEL: ${{ vars.RUNNER_E2E_MAX_PARALLEL || needs.authorize.outputs.max_parallel_default }}
          MAX_PARALLEL_LIMIT: ${{ needs.authorize.outputs.max_parallel_limit }}
+          REQUESTED_MAX_PARALLEL: ${{ inputs.max_parallel }}
          ROSTERS: ${{ inputs.rosters || 'all' }}
        run: |
          set -euo pipefail
@@ -257,6 +262,13 @@ jobs:
            echo "RUNNER_E2E_MAX_PARALLEL must be an integer from 2 through $MAX_PARALLEL_LIMIT for the two-shard direct suite." >&2
            exit 1
          fi
+          if [ -n "${REQUESTED_MAX_PARALLEL:-}" ]; then
+            if ! [[ "$REQUESTED_MAX_PARALLEL" =~ ^[1-9][0-9]{0,2}$ ]] || [ "$REQUESTED_MAX_PARALLEL" -lt 2 ] || [ "$REQUESTED_MAX_PARALLEL" -gt "$MAX_PARALLEL" ]; then
+              echo "max_parallel must be an integer from 2 through the configured campaign limit." >&2
+              exit 1
+            fi
+            MAX_PARALLEL="$REQUESTED_MAX_PARALLEL"
+          fi
          node packages/paperclip-runner/scripts/runner-protocol-eval-campaign.mjs catalog \
            --evals-root .paperclip-evals \
            --rosters "$ROSTERS" \
@@ -332,6 +344,12 @@ jobs:
      - name: Materialize the pinned OpenCode executable before packaging
        run: node packages/paperclip-runner/scripts/materialize-opencode-binary.mjs

+      - name: Materialize the pinned Grok executable when the target includes it
+        run: |
+          if [ -f packages/grok-acp/install.mjs ]; then
+            node packages/grok-acp/install.mjs
+          fi
+
      - name: Build runner CLI, daemon, and canonical attempt viewer
        run: |
          set -euo pipefail
@@ -523,6 +541,7 @@ jobs:
          OPENAI_API_KEY: ${{ matrix.credentialName == 'OPENAI_API_KEY' && secrets.OPENAI_API_KEY || '' }}
          ANTHROPIC_API_KEY: ${{ matrix.credentialName == 'ANTHROPIC_API_KEY' && secrets.ANTHROPIC_API_KEY || '' }}
          OPENROUTER_API_KEY: ${{ matrix.credentialName == 'OPENROUTER_API_KEY' && secrets.OPENROUTER_API_KEY || '' }}
+          XAI_API_KEY: ${{ matrix.credentialName == 'XAI_API_KEY' && secrets.XAI_API_KEY || '' }}
          PAPERCLIP_CLAUDE_MANAGED_PROFILE_ID: ${{ vars.PAPERCLIP_CLAUDE_MANAGED_PROFILE_ID }}
          PAPERCLIP_CLAUDE_MANAGED_AGENT_ID: ${{ vars.PAPERCLIP_CLAUDE_MANAGED_AGENT_ID }}
          PAPERCLIP_CLAUDE_MANAGED_AGENT_VERSION: ${{ vars.PAPERCLIP_CLAUDE_MANAGED_AGENT_VERSION }}
@@ -205,6 +205,7 @@ The paid jobs read only the credential selected for each roster:
 - `OPENAI_API_KEY` for native Codex and ACPX Codex;
 - `ANTHROPIC_API_KEY` for ACPX Claude and Claude Managed;
 - `OPENROUTER_API_KEY` for native OpenCode and ACPX Pi;
+- `XAI_API_KEY` for an explicitly selected ACPX Grok roster;
 - short-lived GitHub OIDC workload identity for AWS AgentCore.

 Claude Managed also requires the four nonsecret
@@ -308,3 +309,14 @@ pnpm --filter @paperclipai/paperclip-runner \
  --campaign-id gha-1-1 \
  --output /tmp/runner-protocol-eval-catalog.json
 ```
+
+
+For Grok qualification, select `protocol-live-acpx-grok` and its exact eval
+revision. Set `max_parallel: 2` when the API key has a low custom rate limit;
+this admits one case per shard. The override can only lower the configured
+campaign ceiling and cannot change the key's provider limits. Retain any
+rate-limited attempts as failures. The build installs the target's pinned,
+checksum-verified Grok binary before packaging the portable runtime. Targets
+without the Grok package keep their existing build behavior. This protocol
+workflow uses the explicitly selected API key; subscription credentials are
+not delivered by it.
@@ -64,6 +64,7 @@ export function credentialForConfig(config) {
  if (config.provider === "acpx") {
    if (config.acpxAgent === "pi") return "OPENROUTER_API_KEY";
    if (config.acpxAgent === "claude") return "ANTHROPIC_API_KEY";
+    if (config.acpxAgent === "grok") return "XAI_API_KEY";
    if (config.acpxAgent === "codex") return "OPENAI_API_KEY";
  }
  throw new Error(
@@ -85,6 +85,7 @@ async function fixture() {

 test("maps every qualified driver to one explicit credential boundary", () => {
  assert.equal(credentialForConfig({ provider: "codex" }), "OPENAI_API_KEY");
+  assert.equal(credentialForConfig({ provider: "acpx", acpxAgent: "grok" }), "XAI_API_KEY");
  assert.equal(
    credentialForConfig({ provider: "opencode" }),
    "OPENROUTER_API_KEY",
@@ -1,4 +1,5 @@
 import assert from "node:assert/strict";
+import { spawnSync } from "node:child_process";
 import { existsSync, readdirSync, readFileSync } from "node:fs";
 import path from "node:path";
 import { fileURLToPath } from "node:url";
@@ -320,7 +321,7 @@ test("Runner eval workflows pin actions and gate paid live execution", () => {
  ];
  const paidWorkflowNameSet = new Set(paidWorkflowNames);
  const providerSecretReference =
-    /secrets(?:\.(?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|DAYTONA_API_KEY)\b|\[['"](?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|DAYTONA_API_KEY)['"]\])/g;
+    /secrets(?:\.(?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|GROK_AUTH_JSON|DAYTONA_API_KEY)\b|\[['"](?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|GROK_AUTH_JSON|DAYTONA_API_KEY)['"]\])/g;
  for (const name of readdirSync(path.join(repoRoot, ".github/workflows"))) {
    if (!/\.ya?ml$/.test(name)) continue;
    const workflow = readWorkflow(name);
@@ -430,3 +431,27 @@ test("Runner eval workflows pin actions and gate paid live execution", () => {
    }
  }
 });
+
+
+test("direct Grok qualification installs the pinned binary and scopes its API key", () => {
+  const workflow = readWorkflow("runner-protocol-live-evals.yml");
+  assert.ok(workflow.includes("XAI_API_KEY: ${{ matrix.credentialName == 'XAI_API_KEY' && secrets.XAI_API_KEY || '' }}"));
+  assert.ok(workflow.includes("if [ -f packages/grok-acp/install.mjs ]; then"));
+  assert.ok(workflow.indexOf("node packages/grok-acp/install.mjs") < workflow.indexOf("pnpm --filter @paperclipai/paperclip-runner deploy --prod"));
+  assert.ok(!workflow.includes("secrets.GROK_AUTH_JSON"));
+});
+
+test("direct protocol concurrency override only lowers the configured ceiling", () => {
+  const workflow = readWorkflow("runner-protocol-live-evals.yml");
+  const start = workflow.indexOf('          if [ -n "${REQUESTED_MAX_PARALLEL:-}" ]; then');
+  const end = workflow.indexOf("          node packages/paperclip-runner/scripts/runner-protocol-eval-campaign.mjs catalog", start);
+  assert.ok(start > 0 && end > start);
+  const script = workflow.slice(start, end) + '\nprintf "%s" "$MAX_PARALLEL"\n';
+  for (const [requested, expected] of [["", "8"], ["2", "2"], ["8", "8"], ["1", null], ["9", null], ["0", null], ["-1", null], ["2.5", null], ["garbage", null], ["9999999999999999999999", null]]) {
+    const result = spawnSync("bash", ["-eu", "-c", script], {
+      env: { ...process.env, MAX_PARALLEL: "8", REQUESTED_MAX_PARALLEL: requested }, encoding: "utf8",
+    });
+    assert.equal(result.status, expected === null ? 1 : 0, requested);
+    if (expected !== null) assert.equal(result.stdout, expected);
+  }
+});
@@ -40,6 +40,14 @@ export interface CampaignBillingSummary {
  testsWithCompleteBilling: number;
 }

+/** Render observed subtotals without presenting absent measurements as zero. */
+export function billingCoverageLabel(value: string, covered: number, total: number): string {
+  if (!Number.isInteger(covered) || !Number.isInteger(total) || covered <= 0 || total <= 0 || covered > total) {
+    return "Unavailable";
+  }
+  return covered < total ? `${value} (partial: ${covered}/${total} runs)` : value;
+}
+
 function record(value: unknown): Record<string, unknown> {
  return value && typeof value === "object" && !Array.isArray(value)
    ? (value as Record<string, unknown>)
@@ -21,6 +21,7 @@ interface PublishedResult extends RunnerE2EResult {
 interface PublishedCampaign {
  schema?: string;
  campaignId?: string;
+  source?: RunnerE2ECampaign["source"];
  generatedAt: string;
  expected: string[];
  results: PublishedResult[];
@@ -94,6 +95,16 @@ export async function regenerateRunnerDashboard(input: {
    expected,
    results,
  });
+  // A rendering refresh must not inherit the renderer's checkout or CI event.
+  const retainedSource = results.find((result) =>
+    result.source?.sha || result.source?.ref || result.source?.workflowRunUrl,
+  )?.source;
+  campaign.source = normalized.source ?? {
+    sha: retainedSource?.sha ?? null,
+    ref: retainedSource?.ref ?? null,
+    workflowRunUrl: retainedSource?.workflowRunUrl ?? null,
+    eventName: null,
+  };
  const history =
    input.historyFile === null
      ? undefined
@@ -5,6 +5,7 @@ import {
  type ReportExecution,
 } from "./report-catalog.js";
 import type {
+  RunnerE2EBillingSummary,
  RunnerE2ECampaign,
  RunnerE2EHistoryIndex,
  RunnerE2EResult,
@@ -12,6 +13,7 @@ import type {
 } from "./types.js";
 import {
  aggregateCampaignBilling,
+  billingCoverageLabel,
  summarizeExecutionBilling,
 } from "./billing.js";

@@ -57,8 +59,13 @@ function durationLabel(durationMs: number) {
  return `${Math.floor(seconds / 60)}m ${seconds % 60}s`;
 }

-function tokenLabel(value: number) {
-  return new Intl.NumberFormat("en-US").format(value);
+function tokenLabel(value: number, llm?: RunnerE2EBillingSummary["llm"]) {
+  const formatted = new Intl.NumberFormat("en-US").format(value);
+  return llm ? billingCoverageLabel(formatted, llm.runsWithTokenUsage, llm.runCount) : formatted;
+}
+
+function llmCostLabel(value: number, llm: RunnerE2EBillingSummary["llm"]) {
+  return billingCoverageLabel(usdLabel(value), llm.runsWithReportedCost, llm.runCount);
 }

 function usdLabel(value: number | null) {
@@ -264,7 +271,7 @@ function renderCase(
            data-gallery-runtime="${html(entry?.result.runtimeMode ?? execution.profile.expectedRuntimeMode)}"
            data-gallery-status="${html(label)}"
            data-gallery-duration="${html(entry ? durationLabel(entry.result.durationMs) : "Not run")}"
-            data-gallery-tokens="${html(billing ? `${tokenLabel(billing.llm.inputTokens)} in · ${tokenLabel(billing.llm.outputTokens)} out` : "Unavailable")}"
+            data-gallery-tokens="${html(billing ? `${tokenLabel(billing.llm.inputTokens, billing.llm)} in · ${tokenLabel(billing.llm.outputTokens, billing.llm)} out` : "Unavailable")}"
            data-gallery-matchers="${html(matcherResults.length > 0 ? `${passedMatchers}/${matcherResults.length} matchers passed${notReached ? ` · ${notReached} not reached` : ""}` : "No matchers recorded")}"
            aria-label="Open ${html(item.label)} for ${html(execution.id)} in gallery"
          >
@@ -276,8 +283,8 @@ function renderCase(
    : "";
  const billingStrip = billing
    ? `<div class="billing-strip" aria-label="Billing for ${html(execution.id)}">
-        <div><span>Tokens</span><strong>${html(tokenLabel(billing.llm.inputTokens))} in · ${html(tokenLabel(billing.llm.outputTokens))} out</strong><small>${html(tokenLabel(billing.llm.cachedInputTokens))} cached · ${billing.llm.runsWithTokenUsage}/${billing.llm.runCount} runs covered</small></div>
-        <div><span>LLM spend</span><strong>${billing.llm.runsWithReportedCost > 0 ? html(usdLabel(billing.reportedCostUsd)) : html(billing.llm.costStatus)}</strong><small>${billing.llm.runsWithReportedCost}/${billing.llm.runCount} runs provider-priced</small></div>
+        <div><span>Tokens</span><strong>${html(tokenLabel(billing.llm.inputTokens, billing.llm))} in · ${html(tokenLabel(billing.llm.outputTokens, billing.llm))} out</strong><small>${html(tokenLabel(billing.llm.cachedInputTokens, billing.llm))} cached · ${billing.llm.runsWithTokenUsage}/${billing.llm.runCount} runs covered</small></div>
+        <div><span>LLM spend</span><strong>${html(llmCostLabel(billing.reportedCostUsd, billing.llm))}</strong><small>${billing.llm.runsWithReportedCost}/${billing.llm.runCount} runs provider-priced</small></div>
        <div><span>Execution</span><strong>${billing.runtime.estimatedListCostUsd === undefined ? html(billing.runtime.costStatus === "not_metered" ? "Local · not metered" : "Cost unavailable") : `${html(usdLabel(billing.runtime.estimatedListCostUsd))} est.`}</strong><small>${html(durationLabel(billing.runtime.agentRunDurationMs))} agent${billing.runtime.leaseDurationMs === null ? "" : ` · ${html(durationLabel(billing.runtime.leaseDurationMs))} lease`}</small></div>
      </div>`
    : "";
@@ -516,8 +523,8 @@ function renderHistory(history: RunnerE2EHistoryIndex | undefined) {
        <td><a href="${html(campaign.publicUrl)}">${html(campaign.campaignId)}</a><small>${html(new Date(campaign.generatedAt).toLocaleString("en-US", { timeZone: "UTC" }))} UTC</small></td>
        <td>${sha ? `<code>${html(sha.slice(0, 10))}</code>` : "Unknown"}<small>${html(campaign.source.ref ?? "unknown ref")}</small></td>
        <td><span class="status history-${status}">${status}</span><small>${campaign.passed}/${campaign.selected} passed${campaign.incomplete ? ` · ${campaign.incomplete} incomplete` : ""} · ${campaign.complete ? "complete" : "partial"}</small></td>
-        <td>${html(tokenLabel(campaign.billing.llm.inputTokens))} / ${html(tokenLabel(campaign.billing.llm.outputTokens))}<small>input / output · ${html(tokenLabel(campaign.billing.llm.cachedInputTokens))} cached</small></td>
-        <td>${html(usdLabel(campaign.billing.reportedLlmCostUsd))}<small>${html(usdLabel(campaign.billing.estimatedRuntimeCostUsd))} runtime estimate</small></td>
+        <td>${html(tokenLabel(campaign.billing.llm.inputTokens, campaign.billing.llm))} / ${html(tokenLabel(campaign.billing.llm.outputTokens, campaign.billing.llm))}<small>input / output · ${html(tokenLabel(campaign.billing.llm.cachedInputTokens, campaign.billing.llm))} cached</small></td>
+        <td>${html(llmCostLabel(campaign.billing.reportedLlmCostUsd, campaign.billing.llm))}<small>${html(usdLabel(campaign.billing.estimatedRuntimeCostUsd))} runtime estimate</small></td>
        <td>${html(durationLabel(campaign.billing.agentRunDurationMs))}<small>${html(durationLabel(campaign.billing.leaseDurationMs))} lease</small></td>
      </tr>`;
    })
@@ -607,8 +614,8 @@ function renderSuiteMatrix(input: {
  const summaryHtml = summary
    ? `<div class="suite-summary" aria-label="${html(suite.label)} current campaign summary">
        <div><span>Pass rate</span><strong>${summary.selected > 0 ? ((summary.passed / summary.selected) * 100).toFixed(1) : "0.0"}%</strong><small>${summary.passed}/${summary.selected} passed${summary.incomplete ? ` · ${summary.incomplete} incomplete` : ""}</small></div>
-        <div><span>Tokens</span><strong>${html(tokenLabel(summary.billing.llm.totalTokens))}</strong><small>${html(tokenLabel(summary.billing.llm.inputTokens))} input · ${html(tokenLabel(summary.billing.llm.outputTokens))} output</small></div>
-        <div><span>Cost</span><strong>${html(usdLabel(summary.billing.observedAndEstimatedCostUsd))}</strong><small>reported LLM + runtime${summary.billing.judge ? " + judge" : ""} estimate</small></div>
+        <div><span>Tokens</span><strong>${html(tokenLabel(summary.billing.llm.totalTokens, summary.billing.llm))}</strong><small>${html(tokenLabel(summary.billing.llm.inputTokens, summary.billing.llm))} input · ${html(tokenLabel(summary.billing.llm.outputTokens, summary.billing.llm))} output</small></div>
+        <div><span>Known spend</span><strong>${html(summary.billing.llm.runsWithReportedCost > 0 || summary.billing.estimatedRuntimeCostUsd > 0 || summary.billing.judge ? `${usdLabel(summary.billing.observedAndEstimatedCostUsd)}${summary.billing.testsWithCompleteBilling < summary.billing.testCount ? " (partial)" : ""}` : "Unavailable")}</strong><small>reported LLM + runtime${summary.billing.judge ? " + judge" : ""} estimate</small></div>
        <div><span>Agent time</span><strong>${html(durationLabel(summary.billing.agentRunDurationMs))}</strong><small>${html(durationLabel(summary.billing.leaseDurationMs))} lease</small></div>
        <div><span>Execution</span><strong>${summary.executed}/${summary.selected}</strong><small>${summary.retries} retries · cleanup ${summary.cleanupPassed ? "passed" : "failed"}</small></div>
      </div>`
@@ -1118,10 +1125,10 @@ export function renderRunnerE2EDashboard(input: RunnerDashboardInput) {
    </header>
    ${publicSummaryImageHref ? `<figure class="public-summary"><img src="${html(publicSummaryImageHref)}" alt="Runner E2E campaign status summary"></figure>` : ""}
    <section class="billing-overview" aria-label="Campaign billing summary">
-      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.inputTokens))}</strong><span>Input tokens</span></div>
-      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.outputTokens))}</strong><span>Output tokens</span></div>
-      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.cachedInputTokens))}</strong><span>Cached tokens</span></div>
-      <div class="billing-metric"><strong>${html(usdLabel(campaignBilling.reportedLlmCostUsd))}</strong><span>LLM reported subtotal</span></div>
+      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.inputTokens, campaignBilling.llm))}</strong><span>Input tokens</span></div>
+      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.outputTokens, campaignBilling.llm))}</strong><span>Output tokens</span></div>
+      <div class="billing-metric"><strong>${html(tokenLabel(campaignBilling.llm.cachedInputTokens, campaignBilling.llm))}</strong><span>Cached tokens</span></div>
+      <div class="billing-metric"><strong>${html(llmCostLabel(campaignBilling.reportedLlmCostUsd, campaignBilling.llm))}</strong><span>LLM reported subtotal</span></div>
      <div class="billing-metric"><strong>${html(usdLabel(campaignBilling.estimatedRuntimeCostUsd))}</strong><span>Daytona list estimate</span></div>
      ${campaignBilling.judge ? `<div class="billing-metric"><strong>${html(usdLabel(campaignBilling.judge.estimatedCostUsd))}</strong><span>Judge estimate · ${campaignBilling.judge.attempts} attempts · ${campaignBilling.judge.attemptsWithUnknownUsage} unknown usage</span><small>${html(tokenLabel(campaignBilling.judge.inputTokens))} in / ${html(tokenLabel(campaignBilling.judge.outputTokens))} out · $${campaignBilling.judge.reservedCostUsd.toFixed(6)} reserved</small></div>` : ""}
      <div class="billing-metric"><strong>${html(durationLabel(campaignBilling.agentRunDurationMs))}</strong><span>Agent execution time</span></div>
@@ -67,6 +67,55 @@ function result(execution: MatrixExecution, status: "passed" | "failed") {
 }

 describe("runner E2E campaign history", () => {
+  it("does not let empty legacy provenance hide a later measured source", async () => {
+    const root = await mkdtemp(path.join(os.tmpdir(), "runner-legacy-source-"));
+    temporaryDirectories.push(root);
+    const source = {
+      sha: "measured-source-sha", ref: "refs/heads/measured-source",
+      workflowRunUrl: "https://example.test/actions/runs/1",
+    };
+    const results = [
+      { ...result(runnerMatrix[0]!, "passed"), schema: "paperclip.runner-e2e.result/v1" },
+      { ...result(runnerMatrix[1]!, "passed"), source },
+    ];
+    await writeFile(path.join(root, "normalized-results.json"), JSON.stringify({
+      campaignId: "legacy-source", generatedAt: results[0]!.finishedAt,
+      expected: results.map((entry) => entry.executionId), results,
+    }));
+    vi.stubEnv("PAPERCLIP_RUNNER_E2E_SOURCE_SHA", "renderer-sha");
+    vi.stubEnv("GITHUB_EVENT_NAME", "push");
+    await regenerateRunnerDashboard({ bundle: root });
+    const regenerated = JSON.parse(await readFile(path.join(root, "normalized-results.json"), "utf8"));
+    expect(regenerated.source).toEqual({ ...source, eventName: null });
+    expect(regenerated.results[0].source).toEqual({ sha: null, ref: null, workflowRunUrl: null });
+    expect(regenerated.passed).toBe(2);
+  });
+
+  it.each(["workflow_dispatch", null, undefined])("preserves retained source during regeneration (event %s)", async (eventName) => {
+    const root = await mkdtemp(path.join(os.tmpdir(), "runner-retained-source-"));
+    temporaryDirectories.push(root);
+    const execution = runnerMatrix[0]!;
+    const source = {
+      sha: "measured-source-sha", ref: "refs/heads/measured-source",
+      workflowRunUrl: "https://example.test/actions/runs/1",
+    };
+    const retainedResult = { ...result(execution, "passed"), source };
+    const campaign = buildRunnerCampaign({
+      campaignId: "retained-source", generatedAt: retainedResult.finishedAt,
+      expected: [execution.id], results: [retainedResult],
+    });
+    const published = { ...campaign, source: eventName === undefined ? undefined : { ...source, eventName } };
+    await writeFile(path.join(root, "normalized-results.json"), JSON.stringify(published));
+    vi.stubEnv("PAPERCLIP_RUNNER_E2E_SOURCE_SHA", "renderer-sha");
+    vi.stubEnv("PAPERCLIP_RUNNER_E2E_SOURCE_REF", "refs/heads/renderer");
+    vi.stubEnv("GITHUB_EVENT_NAME", "push");
+    await regenerateRunnerDashboard({ bundle: root });
+    const regenerated = JSON.parse(await readFile(path.join(root, "normalized-results.json"), "utf8"));
+    expect(regenerated.source).toEqual({ ...source, eventName: eventName ?? null });
+    expect(regenerated.results).toEqual(campaign.results.map((entry) => ({ ...entry, evidenceValid: true, evidenceErrors: [] })));
+    expect(regenerated.generatedAt).toBe(campaign.generatedAt);
+  });
+
  it("retains incomplete journeys in campaign, suite and history without marking them green", () => {
    const execution = runnerMatrix.find(e => e.suite.id === "first-task")!;
    const incomplete: RunnerE2EResult = {
@@ -18,6 +18,49 @@ afterEach(async () => {
 });

 describe("runner E2E report aggregation", () => {
+  it.each([
+    { name: "missing", usage: null, runIds: ["run-1"], tokens: "Unavailable", cost: "Unavailable", htmlTokens: "Unavailable" },
+    { name: "partial", usage: { runs: [{ usage: { inputTokens: 1250, outputTokens: 75, cachedInputTokens: 500, costUsd: 0.0125 } }, { usage: null }] }, runIds: ["run-1", "run-2"], tokens: "1250 input / 75 output / 500 cached (partial: 1/2 runs)", cost: "$0.012500 (partial: 1/2 runs)", htmlTokens: "1,250 (partial: 1/2 runs)" },
+    { name: "reported", usage: { inputTokens: 1250, outputTokens: 75, cachedInputTokens: 500, costUsd: 0.0125 }, runIds: ["run-1"], tokens: "1250 input / 75 output / 500 cached", cost: "$0.012500", htmlTokens: "1,250" },
+    { name: "reported zero cost", usage: { inputTokens: 1, outputTokens: 0, costUsd: 0 }, runIds: ["run-1"], tokens: "1 input / 0 output / 0 cached", cost: "$0.000000", htmlTokens: "1" },
+  ])("renders $name usage with its actual coverage", async ({ usage, runIds, tokens, cost, htmlTokens }) => {
+    const root = await mkdtemp(path.join(os.tmpdir(), "runner-billing-report-"));
+    cleanupDirectories.push(root);
+    const executionId = "core-compatibility.runner-codex.local.message-marker";
+    const directory = path.join(root, "attempt-1");
+    await mkdir(directory);
+    const result: RunnerE2EResult = {
+      schema: "paperclip.runner-e2e.result/v2", executionId, suiteId: "core-compatibility",
+      attempt: 1, status: "passed", profileId: "runner-codex", environmentId: "local",
+      caseId: "message-marker", provider: "codex", model: "fixture-model", runtimeMode: "native",
+      startedAt: "2026-09-23T00:00:00Z", finishedAt: "2026-09-23T00:00:01Z", durationMs: 1000,
+      cleanup: "passed", runIds, usage,
+    };
+    const raw = JSON.stringify(result);
+    await writeFile(path.join(directory, "result.json"), raw);
+    await writeFile(path.join(directory, "final-state.png"), "fake-png");
+    await writeFile(path.join(directory, "evidence-manifest.json"), JSON.stringify({ files: ["final-state.png"], leaks: [], missing: [] }));
+    const output = path.join(root, "merged");
+    await execFileAsync(process.execPath, [path.join(repositoryRoot, "cli/node_modules/tsx/dist/cli.mjs"), path.join(repositoryRoot, "tests/runner-e2e/report.ts")], {
+      cwd: repositoryRoot,
+      env: { ...process.env, PAPERCLIP_RUNNER_E2E_REPORT_ROOT: root, PAPERCLIP_RUNNER_E2E_REPORT_OUT: output, PAPERCLIP_RUNNER_E2E_EXPECTED_IDS: JSON.stringify([executionId]) },
+    });
+    const markdown = await readFile(path.join(output, "summary.md"), "utf8");
+    expect(markdown.split("\n").find((line) => line.startsWith("Tokens: "))).toBe(`Tokens: ${tokens}`);
+    expect(markdown.split("\n").find((line) => line.startsWith("Provider-reported LLM cost: "))).toBe(`Provider-reported LLM cost: ${cost}`);
+    const dashboard = await readFile(path.join(output, "index.html"), "utf8");
+    expect(dashboard).toContain(`<strong>${htmlTokens}</strong><span>Input tokens</span>`);
+    if (usage === null) {
+      expect(markdown).not.toContain("0/0 | $0.000000");
+      expect(dashboard).toContain("<strong>Unavailable</strong><span>LLM reported subtotal</span>");
+      expect(dashboard).toContain("<span>Known spend</span><strong>Unavailable</strong>");
+    }
+    expect(await readFile(path.join(directory, "result.json"), "utf8")).toBe(raw);
+    const normalized = JSON.parse(await readFile(path.join(output, "normalized-results.json"), "utf8"));
+    expect(normalized.passed).toBe(1);
+    expect(normalized.results[0].usage).toEqual(usage);
+  });
+
  it("keeps interrupted journeys incomplete unless their evidence is invalid", async () => {
    const root = await mkdtemp(path.join(os.tmpdir(), "runner-incomplete-report-"));
    cleanupDirectories.push(root);
@@ -8,7 +8,7 @@ import {
  writeFile,
 } from "node:fs/promises";
 import { runnerMatrix } from "./catalog.js";
-import { summarizeExecutionBilling } from "./billing.js";
+import { billingCoverageLabel, summarizeExecutionBilling } from "./billing.js";
 import { renderRunnerE2EDashboard } from "./dashboard.js";
 import {
  buildRunnerCampaign,
@@ -368,9 +368,9 @@ async function main() {
          "",
        ]
      : []),
-    `Tokens: ${billing.llm.inputTokens} input / ${billing.llm.outputTokens} output / ${billing.llm.cachedInputTokens} cached`,
+    `Tokens: ${billingCoverageLabel(`${billing.llm.inputTokens} input / ${billing.llm.outputTokens} output / ${billing.llm.cachedInputTokens} cached`, billing.llm.runsWithTokenUsage, billing.llm.runCount)}`,
    "",
-    `Provider-reported LLM cost: $${billing.reportedLlmCostUsd.toFixed(6)} (${billing.llm.runsWithReportedCost}/${billing.llm.runCount} runs priced)`,
+    `Provider-reported LLM cost: ${billingCoverageLabel(`$${billing.reportedLlmCostUsd.toFixed(6)}`, billing.llm.runsWithReportedCost, billing.llm.runCount)}`,
    "",
    `Estimated Daytona list-price runtime cost: $${billing.estimatedRuntimeCostUsd.toFixed(6)}`,
    ...(billing.judge ? [`Estimated judge cost: ${billing.judge.estimatedCostUsd === null ? "unknown" : `$${billing.judge.estimatedCostUsd.toFixed(6)}`}; ${billing.judge.attempts} attempts; ${billing.judge.attemptsWithUnknownUsage} with unknown usage; $${billing.judge.reservedCostUsd.toFixed(6)} reserved`] : []),
@@ -385,7 +385,7 @@ async function main() {
      const cell = publicCampaignUrl
        ? `[${resolved.executionId}](${publicCampaignUrl}#execution-${encodeURIComponent(resolved.executionId)})`
        : resolved.executionId;
-      return `| ${cell} | ${resolved.attempt} | ${entry.valid ? "pass" : "fail"} | ${resolved.runtimeMode} | ${Math.round(resolved.durationMs / 1000)}s | ${cellBilling.llm.inputTokens}/${cellBilling.llm.outputTokens} | $${cellBilling.reportedCostUsd.toFixed(6)} (${cellBilling.llm.costStatus}) | ${runtimeCost === undefined ? cellBilling.runtime.costStatus : `$${runtimeCost.toFixed(6)} est.`} | ${detail} |`;
+      return `| ${cell} | ${resolved.attempt} | ${entry.valid ? "pass" : "fail"} | ${resolved.runtimeMode} | ${Math.round(resolved.durationMs / 1000)}s | ${billingCoverageLabel(`${cellBilling.llm.inputTokens}/${cellBilling.llm.outputTokens}`, cellBilling.llm.runsWithTokenUsage, cellBilling.llm.runCount)} | ${billingCoverageLabel(`$${cellBilling.reportedCostUsd.toFixed(6)}`, cellBilling.llm.runsWithReportedCost, cellBilling.llm.runCount)} | ${runtimeCost === undefined ? cellBilling.runtime.costStatus : `$${runtimeCost.toFixed(6)} est.`} | ${detail} |`;
    }),
    "",
  ];
@@ -305,7 +305,7 @@ describe("public repository paid workflow security", () => {
    const grokPreparation = paidJob.indexOf("- name: Install checksum-verified Grok executable");
    expect(grokPreparation).toBeGreaterThan(paidInstall);
    expect(paidExecution).toBeGreaterThan(grokPreparation);
-    expect(paidJob).toContain("if: matrix.environmentId == 'local' && matrix.profileId == 'runner-acpx-grok'");
+    expect(paidJob).toContain("if: matrix.environmentId == 'local' && (matrix.profileId == 'runner-acpx-grok' || matrix.profileId == 'runner-acpx-grok-subscription')");
    expect(paidJob).toContain("run: node packages/grok-acp/install.mjs");

    const everydayOracleStep = paidJob.slice(
@@ -313,7 +313,7 @@ describe("public repository paid workflow security", () => {
      paidExecution,
    );
    expect(everydayOracleStep).toContain(
-      "if: (matrix.suiteId == 'everyday-workflows' || matrix.suiteId == 'grok-qualification') && (matrix.caseId == 'build-revise' || matrix.caseId == 'delegate-feedback' || matrix.caseId == 'agent-review-handoff' || matrix.caseId == 'hire-reuse' || matrix.caseId == 'recover-controller' || matrix.caseId == 'stop-redirect')",
+      "if: (matrix.suiteId == 'everyday-workflows' || matrix.suiteId == 'grok-qualification' || matrix.suiteId == 'grok-subscription-qualification') && (matrix.caseId == 'build-revise' || matrix.caseId == 'delegate-feedback' || matrix.caseId == 'agent-review-handoff' || matrix.caseId == 'hire-reuse' || matrix.caseId == 'recover-controller' || matrix.caseId == 'stop-redirect')",
    );
    expect(everydayOracleStep).toContain(
      `oracle_image='${everydayOracleImage}'`,
@@ -530,6 +530,7 @@ describe("public repository paid workflow security", () => {
      ANTHROPIC_API_KEY: "matrix.credentialName == 'ANTHROPIC_API_KEY'",
      OPENROUTER_API_KEY: "matrix.credentialName == 'OPENROUTER_API_KEY'",
      XAI_API_KEY: "matrix.credentialName == 'XAI_API_KEY'",
+      GROK_AUTH_JSON: "matrix.credentialName == 'GROK_AUTH_JSON'",
      DAYTONA_API_KEY: "matrix.environmentId == 'daytona'",
    })) {
      expect(fullStack).toContain(
@@ -557,7 +558,7 @@ describe("public repository paid workflow security", () => {
      );
      const providerSecretReferences = [
        ...contents.matchAll(
-          /secrets(?:\.(?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|DAYTONA_API_KEY)\b|\[['"](?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|DAYTONA_API_KEY)['"]\])/g,
+          /secrets(?:\.(?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|GROK_AUTH_JSON|DAYTONA_API_KEY)\b|\[['"](?:OPENAI_API_KEY|ANTHROPIC_API_KEY|OPENROUTER_API_KEY|XAI_API_KEY|GROK_AUTH_JSON|DAYTONA_API_KEY)['"]\])/g,
        ),
      ];
      if (providerSecretReferences.length > 0) {