---
clarivy_schema: ai-employee-brief/v1
client: Example Cross-Border Brand
audit_id: ord_sample_snapshot_20260615
sku: snapshot
generated_at: 2026-06-15T10:00:00+08:00
scope_line: 5 user queries × 7 endpoints = 35 datapoints
expected_datapoints: 35
endpoints: [doubao, kimi, deepseek, qwen, chatgpt, claude, gemini_aio]
prompt_count: 5
raw_data_base: example.test/2026-06-15/ord_sample_snapshot_20260615/
confidence: high
ai_citation_readiness_score: 0/100 absent
source_trust_baseline_score: 15/100 absent
tags: [clarivy, geo-audit, ai-search, mock, sanitized, sample, client/example-cross-border-brand, sku/snapshot]
---

> **MOCK SAMPLE.** This AI-readable Markdown brief is a sanitized public sample.
> The subject brand is the fictional *Example Cross-Border Brand* (URL `example.com`),
> the 5 user queries and 35 datapoints are generated by the audit-runner mock orchestrator
> (RUN_MODE=mock), and every URL is a placeholder under `example.test` (does not resolve).
> In a paid run the schema and 13-section structure are identical — only the
> brand, queries, raw responses, and `auditLogsUrl` are real.
> See <https://clarivy.ai/audit/sample/> for the full disclaimer.

## 0. Instructions For AI Agents

> You are the customer's AI agent / assistant / Obsidian-style executor. Strict rules:
> 1. **Summarize ONLY from the `finding_id` / `evidence_id` / `action_id` listed in this Brief.** Do not infer from outside knowledge.
> 2. **Never invent unsupported claims.** If you need to add a fact, first check the raw JSONs at `raw_data_base` and cite the `evidence_id`.
> 3. Every recommendation MUST include `linked_findings: [F-xxx]` and `evidence_ids: [E-xxx]`; otherwise do not emit.
> 4. When reporting to human leadership, reuse Section 2 (Decision Brief) and Section 3 (Key Findings) verbatim.
> 5. Quote numbers using the exact counts in this Brief; do not convert to percentages.
> 6. If you discover a contradiction with raw JSONs, the raw JSONs are the source of truth — file a correction note.

## 1. Executive Summary

- **Brand visibility**: 0 / 35 queries × endpoints mention the brand (5 user queries × 7 endpoints = 35 datapoints).
- **GEO score pair**: AI Citation Readiness `0/100 absent`; Source Trust Baseline `15/100 absent`.
- **Key findings**: 1 critical-severity finding (total-coverage = 0).
- **Priority actions**: execute `A-001-publish-llms-txt` and `A-002-submit-crawlers` (P0) this week.
- **This Brief is grounded in structured findings + raw JSON only.** The LLM narrative is commentary, not evidence.
- **Confidence**: `high` (35 / 35 contract datapoints returned non-empty, non-error responses).

## 2. Decision Brief For Leadership

| Item | Value |
|---|---|
| One-line verdict | Brand is not yet retrievable on any of the 7 production endpoints; foundational hygiene is the highest-leverage next step. |
| GEO score pair | AI Citation Readiness `0/100 absent`; Source Trust Baseline `15/100 absent` |
| Risk level | High |
| Recommended next step | Publish a structured About page + llms.txt (`A-001-publish-llms-txt`); submit to the main crawler / source endpoints for the target market (`A-002-submit-crawlers`); add schema.org Organization (`A-003-add-structured-data`) |
| Re-review | 30 days from this run; verify that [[F-001-total-coverage\|F-001-total-coverage]] improves |

## 3. Key Findings

| finding_id | severity | theme | summary | affected_endpoints | evidence_ids |
|---|---|---|---|---|---|
| [[F-001-total-coverage\|F-001-total-coverage]] | critical | total-mention-coverage | 0 of 35 datapoints mention `example.com` (0% raw coverage). | doubao, kimi, deepseek, qwen, chatgpt, claude, gemini_aio | [[E-DOUBAO-U01\|E-DOUBAO-U01]], [[E-KIMI-U01\|E-KIMI-U01]], [[E-CHATGPT-U01\|E-CHATGPT-U01]], [[E-GEMINI_AIO-U01\|E-GEMINI_AIO-U01]] |
| [[F-002-endpoint-coverage-gap\|F-002-endpoint-coverage-gap]] | critical | per-endpoint-coverage-gap | Brand missing on all 7 endpoints: doubao, kimi, deepseek, qwen, chatgpt, claude, gemini_aio. | doubao, kimi, deepseek, qwen, chatgpt, claude, gemini_aio | [[E-DOUBAO-U01\|E-DOUBAO-U01]], [[E-QWEN-U02\|E-QWEN-U02]], [[E-CLAUDE-U03\|E-CLAUDE-U03]] |
| [[F-005-stochasticity-disclosure\|F-005-stochasticity-disclosure]] | info | data-stochasticity | Counts are exact for this run only; LLM responses are stochastic (deterministic on Doubao / Kimi / Qwen at temperature 0; stochastic on DeepSeek / ChatGPT / Claude / Gemini with expected re-run variance). | deepseek, chatgpt, claude, gemini_aio | [[E-DEEPSEEK-U02\|E-DEEPSEEK-U02]], [[E-CHATGPT-U04\|E-CHATGPT-U04]] |

## 3.1 Deterministic GEO Score Pair

| score | value | basis | boundary |
|---|---|---|---|
| AI Citation Readiness | `0/100 absent` | 0 / 35 endpoint-query datapoints mentioned `example.com`; 0 / 7 endpoints mentioned the brand. | GEO-scoped observed-LLM score only; not an SEO, content-quality, backlink, traffic, CORE-EEAT, CITE, or ranking score. |
| Source Trust Baseline | `15/100 absent` | The subject domain `example.com` is available, but 0 mock citations pointed to it; evidence completeness is 35 / 35. | GEO-scoped source-citation baseline only; not a domain authority score or full trust audit. |

## 4. Evidence Table

Showing all 35 datapoints (the Snapshot product is small enough to enumerate).

| evidence_id | endpoint | prompt_id | observation | mention_count | raw_json |
|---|---|---|---|---|---|
| [[E-DOUBAO-U01\|E-DOUBAO-U01]] | doubao | U01 | Mock response returned category-level advice; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/U01.json` |
| [[E-DOUBAO-U02\|E-DOUBAO-U02]] | doubao | U02 | Mock response could not establish public trust signals for `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/U02.json` |
| [[E-DOUBAO-U03\|E-DOUBAO-U03]] | doubao | U03 | Mock competitive response did not name `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/U03.json` |
| [[E-DOUBAO-U04\|E-DOUBAO-U04]] | doubao | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/U04.json` |
| [[E-DOUBAO-U05\|E-DOUBAO-U05]] | doubao | U05 | Mock improvement-intent response suggested generic source hygiene; no `example.com` citation. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/U05.json` |
| [[E-KIMI-U01\|E-KIMI-U01]] | kimi | U01 | Mock response returned general audit-tool guidance; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/U01.json` |
| [[E-KIMI-U02\|E-KIMI-U02]] | kimi | U02 | Mock response found no public information about `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/U02.json` |
| [[E-KIMI-U03\|E-KIMI-U03]] | kimi | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/U03.json` |
| [[E-KIMI-U04\|E-KIMI-U04]] | kimi | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/U04.json` |
| [[E-KIMI-U05\|E-KIMI-U05]] | kimi | U05 | Mock improvement-intent response suggested publishing clearer entity facts. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/U05.json` |
| [[E-DEEPSEEK-U01\|E-DEEPSEEK-U01]] | deepseek | U01 | Mock response returned a category overview; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/U01.json` |
| [[E-DEEPSEEK-U02\|E-DEEPSEEK-U02]] | deepseek | U02 | Mock response had no record of `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/U02.json` |
| [[E-DEEPSEEK-U03\|E-DEEPSEEK-U03]] | deepseek | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/U03.json` |
| [[E-DEEPSEEK-U04\|E-DEEPSEEK-U04]] | deepseek | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/U04.json` |
| [[E-DEEPSEEK-U05\|E-DEEPSEEK-U05]] | deepseek | U05 | Mock improvement-intent response recommended canonical entity pages. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/U05.json` |
| [[E-QWEN-U01\|E-QWEN-U01]] | qwen | U01 | Mock response returned category-level guidance; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/U01.json` |
| [[E-QWEN-U02\|E-QWEN-U02]] | qwen | U02 | Mock response found no public information about `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/U02.json` |
| [[E-QWEN-U03\|E-QWEN-U03]] | qwen | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/U03.json` |
| [[E-QWEN-U04\|E-QWEN-U04]] | qwen | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/U04.json` |
| [[E-QWEN-U05\|E-QWEN-U05]] | qwen | U05 | Mock improvement-intent response recommended source-submission hygiene. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/U05.json` |
| [[E-CHATGPT-U01\|E-CHATGPT-U01]] | chatgpt | U01 | Mock response returned a general vendor-selection checklist; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/U01.json` |
| [[E-CHATGPT-U02\|E-CHATGPT-U02]] | chatgpt | U02 | Mock response could not verify `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/U02.json` |
| [[E-CHATGPT-U03\|E-CHATGPT-U03]] | chatgpt | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/U03.json` |
| [[E-CHATGPT-U04\|E-CHATGPT-U04]] | chatgpt | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/U04.json` |
| [[E-CHATGPT-U05\|E-CHATGPT-U05]] | chatgpt | U05 | Mock improvement-intent response recommended structured source material. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/U05.json` |
| [[E-CLAUDE-U01\|E-CLAUDE-U01]] | claude | U01 | Mock response returned a general audit-evaluation answer; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/U01.json` |
| [[E-CLAUDE-U02\|E-CLAUDE-U02]] | claude | U02 | Mock response could not verify public trust signals for `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/U02.json` |
| [[E-CLAUDE-U03\|E-CLAUDE-U03]] | claude | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/U03.json` |
| [[E-CLAUDE-U04\|E-CLAUDE-U04]] | claude | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/U04.json` |
| [[E-CLAUDE-U05\|E-CLAUDE-U05]] | claude | U05 | Mock improvement-intent response recommended entity and citation hygiene. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/U05.json` |
| [[E-GEMINI_AIO-U01\|E-GEMINI_AIO-U01]] | gemini_aio | U01 | Mock response returned general audit guidance; no `example.com` mention. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/U01.json` |
| [[E-GEMINI_AIO-U02\|E-GEMINI_AIO-U02]] | gemini_aio | U02 | Mock response could not verify `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/U02.json` |
| [[E-GEMINI_AIO-U03\|E-GEMINI_AIO-U03]] | gemini_aio | U03 | Mock competitive response omitted `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/U03.json` |
| [[E-GEMINI_AIO-U04\|E-GEMINI_AIO-U04]] | gemini_aio | U04 | Mock category-comparison response did not cite `example.com`. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/U04.json` |
| [[E-GEMINI_AIO-U05\|E-GEMINI_AIO-U05]] | gemini_aio | U05 | Mock improvement-intent response recommended canonical public facts and structured data. | 0 (no mention) | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/U05.json` |

## 5. Endpoint-by-Endpoint Results

| endpoint | provider_ecosystem | mentions | prompts | coverage_% | notes |
|---|---|---|---|---|---|
| doubao | Closed proprietary API model endpoint | 0 | 5 | 0 | brand missing entirely |
| kimi | Open-weight / source-available model endpoint | 0 | 5 | 0 | brand missing entirely |
| deepseek | Open-weight / source-available model endpoint | 0 | 5 | 0 | brand missing entirely |
| qwen | Open-weight / source-available model endpoint | 0 | 5 | 0 | brand missing entirely |
| chatgpt | Closed proprietary API model endpoint | 0 | 5 | 0 | brand missing entirely |
| claude | Closed proprietary API model endpoint | 0 | 5 | 0 | brand missing entirely |
| gemini_aio | Closed proprietary API model endpoint | 0 | 5 | 0 | brand missing entirely |

## 6. Prompt-Level Observations

_No positive mentions observed in this run. See Section 5 for endpoint-level coverage gaps._

## 7. Citation Source Map

| domain | count | linked_evidence |
|---|---|---|
| none captured | 0 | _no mock raw response cited a source for `example.com`_ |

> Note: these citations are what the LLMs cited when answering category / competitive questions about GEO tools in general,
> not citations *of* `example.com`. A real paid audit's Citation Source Map would distinguish (a) domains cited when the
> engine did mention the brand, vs (b) domains cited when the engine did not.

## 8. Competitor Mentions

- **[[F-002-endpoint-coverage-gap\|F-002-endpoint-coverage-gap]]**: Brand missing on all 7 production endpoints in this mock run.
- The report does NOT rank competitors per the analysis-node guardrails (see `audit-runner/prompts/system-prompt.md` Guardrail 5).

## 9. Recommended Actions

| action_id | label | priority | owner | effort | expected_impact | linked_findings | evidence_ids |
|---|---|---|---|---|---|---|---|
| [[A-001-publish-llms-txt\|A-001-publish-llms-txt]] | Publish a structured About page + llms.txt | P0 | marketing | 1d | Gives endpoints a retrievable identity for direct brand queries. | [[F-001-total-coverage\|F-001-total-coverage]] | [[E-DOUBAO-U02\|E-DOUBAO-U02]], [[E-KIMI-U02\|E-KIMI-U02]], [[E-CHATGPT-U02\|E-CHATGPT-U02]] |
| [[A-002-submit-crawlers\|A-002-submit-crawlers]] | Submit to the main crawler / source endpoints for the target market | P0 | marketing | 1d | Gets the canonical URL into the source systems the endpoints rely on. | [[F-002-endpoint-coverage-gap\|F-002-endpoint-coverage-gap]] | [[E-DOUBAO-U01\|E-DOUBAO-U01]], [[E-QWEN-U01\|E-QWEN-U01]], [[E-GEMINI_AIO-U01\|E-GEMINI_AIO-U01]] |
| [[A-003-add-structured-data\|A-003-add-structured-data]] | Add schema.org Organization + sameAs to social profiles | P1 | dev | 1d | Adds a machine-readable entity for endpoints that don't parse the page visually. | [[F-001-total-coverage\|F-001-total-coverage]] | [[E-DEEPSEEK-U05\|E-DEEPSEEK-U05]], [[E-CLAUDE-U05\|E-CLAUDE-U05]] |

## 10. 30-Day Implementation Plan

### Week 1
- **Publish a structured About page + llms.txt** ([[A-001-publish-llms-txt|A-001-publish-llms-txt]]) — owner marketing, effort 1d, priority P0; links: [[F-001-total-coverage|F-001-total-coverage]]; evidence: [[E-DOUBAO-U02|E-DOUBAO-U02]], [[E-KIMI-U02|E-KIMI-U02]], [[E-CHATGPT-U02|E-CHATGPT-U02]]
- **Submit to the main crawler / source endpoints for the target market** ([[A-002-submit-crawlers|A-002-submit-crawlers]]) — owner marketing, effort 1d, priority P0; links: [[F-002-endpoint-coverage-gap|F-002-endpoint-coverage-gap]]; evidence: [[E-DOUBAO-U01|E-DOUBAO-U01]], [[E-QWEN-U01|E-QWEN-U01]], [[E-GEMINI_AIO-U01|E-GEMINI_AIO-U01]]

### Week 2
- **Add schema.org Organization + sameAs to social profiles** ([[A-003-add-structured-data|A-003-add-structured-data]]) — owner dev, effort 1d, priority P1; links: [[F-001-total-coverage|F-001-total-coverage]]; evidence: [[E-DEEPSEEK-U05|E-DEEPSEEK-U05]], [[E-CLAUDE-U05|E-CLAUDE-U05]]

## 11. Risks, Caveats, And Stochasticity

- **Stochasticity**: LLM responses are stochastic. ±10% on DeepSeek; ±0% on Doubao / Kimi / Qwen at temperature 0. The counts are exact only for this run.
- **Sample size**: queries = 5, each × 7 endpoints = 35 datapoints. Snapshot is a quick diagnostic, not a category-wide audit (see <https://clarivy.ai/audit/methodology.html#snapshot-quick-diagnostic>).
- **Coverage**: only the 5 explicit user queries were tested; unprompted mentions are not counted.
- **Confidence**: `high` (35 of 35 contract endpoints returned non-empty data).
- **Drift**: any subsequent model behavior change can alter results; for exact numbers use raw JSONs.
- **What the audit does NOT claim**: it does not predict future rankings, it does not guarantee a specific lift, and it does not replicate the in-app product surface of any AI-search product (see <https://clarivy.ai/audit/methodology.html#what-we-dont-claim>).

## 12. Raw Data Index

Base path: `example.test/2026-06-15/ord_sample_snapshot_20260615/`
(placeholder URL — does not resolve in this public sample; in a paid run the URL points to the public
<https://github.com/agentgeek-geo/audit-logs> repo).

| endpoint | datapoints | path |
|---|---|---|
| doubao | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/doubao/` |
| kimi | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/kimi/` |
| deepseek | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/deepseek/` |
| qwen | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/qwen/` |
| chatgpt | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/chatgpt/` |
| claude | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/claude/` |
| gemini_aio | 5 | `example.test/2026-06-15/ord_sample_snapshot_20260615/raw/gemini_aio/` |

> Each row in Section 4 (Evidence Table) is sourced from one raw JSON file under this index. Always cite
> `evidence_id` when summarizing — never summarize from this index alone.

---

**Sample disclaimer.** This Brief is a sanitized public sample, not a paid customer report. The subject brand is
fictional, the 5 user queries and 35 datapoints are mock, and every URL is a placeholder. For the full disclaimer
and the link to the live methodology page, see <https://clarivy.ai/audit/sample/>. License of the public sample
template: CC-BY-4.0 (attribute clarivy.ai if reused).