💬 GEO Citation Block Draft
Drafts a short answer-engine citation block (claim plus source placeholders) for one blog section, without inventing URLs or statistics — placeholders only.
The prompt
You help an editor draft a short citation block meant for answer engines and careful readers. One blog section only. Do not invent sources, URLs, statistics or legal conclusions.
Article title: {{article_title}}
Section heading: {{section_heading}}
Section draft (claims I want supported):
{{section_draft}}
Audience / region: {{audience}}
Topic sensitivity (e.g. travel tips, product care, compliance-adjacent): {{sensitivity}}
Return:
1. A 2–4 sentence "answer block" that restates only claims already present in my section draft, in plain language a model could quote.
2. A bullet list of citation placeholders, each as:
- Claim: <short claim from my draft>
Source needed: {{source_type_hint}} — use a placeholder like [OFFICIAL_SOURCE_TITLE] and [SOURCE_URL] — never a real URL you invent.
What to verify: one line on what I must check before publishing.
3. If sensitivity mentions compliance, regulation, tax, medical or legal topics, add a one-line disclaimer: this block is editorial drafting help, not legal advice, and must be reviewed by a qualified person before use.
4. Up to 3 follow-up questions I should answer in the article so the citation block stays honest (no new facts — questions only).
Rules:
- Never fabricate statistics, dates, agency names or URLs. If my draft already names a body (e.g. NPS), you may keep that name but still use [SOURCE_URL] placeholders.
- If a claim in my draft is too strong to support, rewrite the answer block to a softer wording and mark the claim as "needs source or cut".
- Do not add SEO keyword stuffing.
Variables
Replace each {{name}} before you send the prompt.
| Variable | What to put in | Example |
|---|---|---|
{{article_title}} | Full article title. | Pre-Shipment Inspection AQL Sampling — What the Report Actually Means |
{{section_heading}} | The H2/H3 this block will sit under. | What AQL 2.5 means on a major defect |
{{section_draft}} | Paste the section text that contains the claims. | On many consumer-goods orders, buyers and inspectors agree an AQL for major defects such as 2.5 under a sampling standard the contract names. The report should state the sample size and how many majors were found. Failing majors usually trigger rework or a CAPA discussion before shipment. Exact acceptance numbers come from the plan on the order — not from a blog post. |
{{audience}} | Who reads this. | US importers new to third-party inspection |
{{sensitivity}} | How sensitive the topic is. | compliance-adjacent quality process; not legal advice |
{{source_type_hint}} | What kind of sources you expect to attach later. | named sampling standard on the PO; inspector's report fields; your own contract clause |
Example
article_title: Pre-Shipment Inspection AQL Sampling — What the Report Actually Means section_heading: What AQL 2.5 means on a major defect section_draft: (as in the example above) audience: US importers new to third-party inspection sensitivity: compliance-adjacent quality process; not legal advice source_type_hint: sampling standard named on the PO; inspection report; contract clause
### 1. Answer Block For many consumer goods orders, buyers and inspectors set an Acceptance Quality Limit (AQL), such as 2.5 for major defects, governed by a sampling standard defined in the contract. The inspection report must document the inspected sample size and the exact count of major defects identified. Failing to meet the threshold for major defects generally prompts rework or a Corrective and Preventive Action (CAPA) discussion prior to shipment, with all definitive acceptance and rejection thresholds established by the specific order’s plan. --- ### 2. Citation Placeholders - **Claim:** Buyers and inspectors agree on an AQL for major defects (e.g., 2.5) governed by a named sampling standard. - **Source needed:** [OFFICIAL_SAMPLING_STANDARD] referenced in your contract/PO at [SOURCE_URL] - **What to verify:** Confirm that your purchase order or quality agreement explicitly names an established standard (such as ANSI/ASQ Z1.4 or ISO 2859-1) rather than assuming a default standard. - **Claim:** The inspection report should state the sample size and how many major defects were found. - **Source needed:** [INSPECTION_REPORT_STANDARD_OR_TEMPLATE] at [SOURCE_URL] - **What to verify:** Review standard reporting templates from your third-party inspection agency to confirm that sample size ($n$) and defect counts per classification are mandatory fields. - **Claim:** Failing majors usually trigger rework or a CAPA discussion before shipment. - **Source needed:** [SUPPLIER_QUALITY_AGREEMENT_OR_SOP] at [SOURCE_URL] - **What to verify:** Check your internal standard operating procedures and vendor terms to ensure they explicitly mandate rework, reinspection, or CAPA protocols upon a failed inspection report. - **Claim:** Exact acceptance numbers come from…
Real, unedited output from a test run on Nov 3, 2026 · Gemini · model gemini-flash-latest.
Full output · Gemini (gemini-flash-latest)
### 1. Answer Block For many consumer goods orders, buyers and inspectors set an Acceptance Quality Limit (AQL), such as 2.5 for major defects, governed by a sampling standard defined in the contract. The inspection report must document the inspected sample size and the exact count of major defects identified. Failing to meet the threshold for major defects generally prompts rework or a Corrective and Preventive Action (CAPA) discussion prior to shipment, with all definitive acceptance and rejection thresholds established by the specific order’s plan. --- ### 2. Citation Placeholders - **Claim:** Buyers and inspectors agree on an AQL for major defects (e.g., 2.5) governed by a named sampling standard. - **Source needed:** [OFFICIAL_SAMPLING_STANDARD] referenced in your contract/PO at [SOURCE_URL] - **What to verify:** Confirm that your purchase order or quality agreement explicitly names an established standard (such as ANSI/ASQ Z1.4 or ISO 2859-1) rather than assuming a default standard. - **Claim:** The inspection report should state the sample size and how many major defects were found. - **Source needed:** [INSPECTION_REPORT_STANDARD_OR_TEMPLATE] at [SOURCE_URL] - **What to verify:** Review standard reporting templates from your third-party inspection agency to confirm that sample size ($n$) and defect counts per classification are mandatory fields. - **Claim:** Failing majors usually trigger rework or a CAPA discussion before shipment. - **Source needed:** [SUPPLIER_QUALITY_AGREEMENT_OR_SOP] at [SOURCE_URL] - **What to verify:** Check your internal standard operating procedures and vendor terms to ensure they explicitly mandate rework, reinspection, or CAPA protocols upon a failed inspection report. - **Claim:** Exact acceptance numbers come from the plan on the order, not a generic source. - **Source needed:** [SPECIFIC_PO_OR_ORDER_SPECIFICATION] at [SOURCE_URL] - **What to verify:** Ensure your purchasing documentation binds the inspection to a specific lot size, inspection level (e.g., General Level II), and AQL table to determine the precise accept/reject ($Ac/Re$) numbers. --- ### 3. Disclaimer This block is editorial drafting help, not legal advice, and must be reviewed by a qualified person before use. --- ### 4. Follow-up Questions 1. Which specific sampling standard (e.g., ANSI/ASQ Z1.4, ISO 2859-1) is officially named in your purchase orders to establish the AQL 2.5 table? 2. How does the contract specifically define what constitutes a "major defect" versus a "minor" or "critical" defect for your product type? 3. Who holds final authority to sign off on a waiver, rework plan, or shipment release if the major defect count exceeds the order's acceptance limit?
Full output · DeepSeek (deepseek-flash)
**Answer block** On many consumer-goods orders, buyers and inspectors agree an AQL for major defects, such as 2.5, under a sampling standard the contract names. The report should state the sample size and how many majors were found. If the major-defect result fails the plan, the next step may be a rework or CAPA discussion before shipment. Exact acceptance numbers come from the plan on the order, not from a blog post. **Citation placeholders** - Claim: On many consumer-goods orders, buyers and inspectors agree an AQL for major defects such as 2.5 under a sampling standard the contract names. Source needed: Named sampling standard on the PO and your contract clause — [OFFICIAL_SOURCE_TITLE], [SOURCE_URL]. What to verify: Confirm the order documents name the standard and AQL, and that “many” is supportable or soften it. - Claim: The report should state the sample size and number of major defects found. Source needed: Inspector’s report fields/template — [OFFICIAL_SOURCE_TITLE], [SOURCE_URL]. What to verify: Confirm the report template has explicit fields for sample size and major-defect count. - Claim: Failing majors may lead to rework or a CAPA discussion before shipment (draft says “usually trigger” — needs source or cut). Source needed: Your own contract clause or inspection protocol — [OFFICIAL_SOURCE_TITLE], [SOURCE_URL]. What to verify: Confirm the contract/protocol states the post-failure step; otherwise soften or cut “usually trigger.” - Claim: Exact acceptance numbers come from the sampling plan on the order, not from a blog post. Source needed: Sampling plan on the PO/order and your contract clause — [OFFICIAL_SOURCE_TITLE], [SOURCE_URL]. What to verify: Confirm the order’s sampling plan contains the acceptance/rejection numbers and is the version cited. **Sensitivity disclaimer** This block is editorial drafting help, not legal advice, and must be reviewed by a qualified person before use. **Follow-up questions** 1. Which sampling standard and AQL does your order/PO actually name for major defects, and where is that stated? 2. Which report fields record sample size and major-defect count, and does your template include them? 3. Does your contract or inspection protocol specify rework, CAPA, or another step after failed majors before shipment?
How to use it
Works in ChatGPT, Claude, Gemini and Perplexity. You get a quotable answer block plus source placeholders — never invented links or stats.
If the section touches compliance, the prompt forces a not-legal-advice line so you do not publish a citation block that sounds like counsel.
Test log
| Date (UTC) | Tool | Model | Run | Result | Checks | Tokens in / out |
|---|---|---|---|---|---|---|
| Nov 3, 2026 | Gemini | gemini-flash-latest | Test | Passed | 3/3
| 0 / 0 |
| Nov 3, 2026 | DeepSeek | deepseek-flash | Test | Passed | 3/3
| 503 / 4,526 |
Each row is a real run recorded by our test runner. For each tool, the latest test run decides whether it appears under “Tested on”. Baseline runs do the same task with no skill installed and never count.