sincLLM operator guide · input contract
AI Search Setup Input Contract: Required Fields, Rejection Rules, and Handoff
Define the minimum input record and deterministic rejection rules before search and AI-crawler discoverability begins.
The direct answer
Define the minimum input record and deterministic rejection rules before search and AI-crawler discoverability begins. The working output is A versioned input-contract table with required fields, validation rules, owners, and rejected-example fixtures.
For AI Search Setup, the bounded capability is search and AI-crawler discoverability. Begin only when the team can supply domain access and a current inventory of the pages that should be discoverable. The documented delivery target is an indexing, schema, and llms.txt setup for the site; anything broader requires a new scope and a new authority decision.
The copyable input contract
This input contract is for site owners whose important pages are missing, inconsistently described, or difficult for crawlers to discover. It begins with domain access and a current inventory of the pages that should be discoverable and stays inside the documented workflow: crawlability, canonical URLs, sitemap discovery, structured data, indexing submission, and a readable llms.txt surface. For AI Search Setup, the input contract remains reviewable because its decisions have named owners, evidence fields, and stop conditions.
Copy this AI Search Setup table into an intake form or machine-readable schema. Its validation column answers whether an input is usable for search and AI-crawler discoverability; its rejection column prevents an incomplete record from entering execution as though it were approved.
| Field | Purpose | Validation rule | Owner | Rejection behavior |
|---|---|---|---|---|
request_id | A stable identifier for this bounded request | Non-empty and unique within the run | site owner | Reject duplicate or missing IDs |
intended_outcome | Define the minimum input record and deterministic rejection rules before search and AI-crawler discoverability begins. | Names one observable decision or artifact | site owner | Reject broad or outcome-guaranteeing language |
input_boundary | domain access and a current inventory of the pages that should be discoverable | Source, owner, freshness, and permitted use are recorded | site owner | Hold when access or provenance is absent |
workflow_scope | crawlability, canonical URLs, sitemap discovery, structured data, indexing submission, and a readable llms.txt surface | Every included stage is named; exclusions stay visible | release reviewer | Reject silent scope expansion |
acceptance_evidence | target URLs are fetchable without an unintended block, canonical links resolve to the intended URLs, sitemap entries match the canonical inventory, structured data parses and agrees with visible content, and the public llms.txt files expose the intended routes | Each criterion maps to an observable check | release reviewer | Return NOT_TESTED when the check cannot run |
failure_fixtures | blocked or contradictory crawl directives, canonical links that point away from the intended page, schema that parses but misdescribes the visible page, orphaned pages absent from internal navigation, and treating llms.txt as a substitute for useful content | At least one safe negative case exists | release reviewer | Reject a success-only test set |
handoff | Owner: release reviewer; deliverable: an indexing, schema, and llms.txt setup for the site | Recipient, format, expiry, and reopen trigger are explicit | release reviewer | Do not release an ownerless artifact |
Example record
{
"contract_version": "1.0",
"request_id": "ART-01-01-EXAMPLE",
"intended_outcome": "Define the minimum input record and deterministic rejection rules before search and AI-crawler discoverability begins.",
"input_boundary": "domain access and a current inventory of the pages that should be discoverable",
"authority": "named owner approval required for consequences outside this artifact",
"acceptance_status": "NOT_TESTED",
"reopen_if": "blocked or contradictory crawl directives"
}
Contract decision
A record is admitted only when every required field is present, its source is named, and the release reviewer can run the associated check. It is held when a missing fact could be supplied without changing scope. It is rejected when the requested effect exceeds the authority of the recorded owner or asks this product to promise an outcome outside its boundary.
Run the workflow as a sequence of decisions
The AI Search Setup input contract follows this working sequence: crawlability, canonical URLs, sitemap discovery, structured data, indexing submission, and a readable llms.txt surface. Within this artifact, each phrase marks a state boundary for search and AI-crawler discoverability. A stage output becomes the next named input, while a failed, missing, or unavailable check keeps the dependent input contract decision closed.
| Step | Decision owner | Observable criterion | Evidence to retain | Counterexample policy |
|---|---|---|---|---|
| 1 | site owner | Target URLs are fetchable without an unintended block. | Direct observation or test bound to the current artifact | Run a safe negative fixture from the separate failure register; do not infer a one-to-one mapping by list position. |
| 2 | search implementation owner | Canonical links resolve to the intended URLs. | Direct observation or test bound to the current artifact | Run a safe negative fixture from the separate failure register; do not infer a one-to-one mapping by list position. |
| 3 | content owner | Sitemap entries match the canonical inventory. | Direct observation or test bound to the current artifact | Run a safe negative fixture from the separate failure register; do not infer a one-to-one mapping by list position. |
| 4 | release reviewer | Structured data parses and agrees with visible content. | Direct observation or test bound to the current artifact | Run a safe negative fixture from the separate failure register; do not infer a one-to-one mapping by list position. |
| 5 | site owner | The public llms.txt files expose the intended routes. | Direct observation or test bound to the current artifact | Run a safe negative fixture from the separate failure register; do not infer a one-to-one mapping by list position. |
Separate failure register
FAIL-01: Blocked or contradictory crawl directives.FAIL-02: Canonical links that point away from the intended page.FAIL-03: Schema that parses but misdescribes the visible page.FAIL-04: Orphaned pages absent from internal navigation.FAIL-05: Treating llms.txt as a substitute for useful content.
The register supplies negative cases for the complete acceptance set. A reviewer determines affected checks from observed evidence; array position never asserts that one failure proves or disproves one criterion.
The producer can explain what it attempted, but the release reviewer evaluates the evidence. If the artifact changes, its prior verdict expires. This is especially important for search and AI-crawler discoverability, where a plausible narrative can hide a stale configuration, an untested negative case, or an authority mismatch.
Failure and recovery drills
A useful AI Search Setup input contract explains what happens when its happy path breaks. These drills come from the accepted product truth record rather than a claim that every buyer has each failure. Use safe synthetic or authorized observations for search and AI-crawler discoverability, and keep private credentials out of every fixture.
1. Blocked or contradictory crawl directives.
Detect for AI Search Setup: site owner captures a direct readback or safe fixture that makes this input contract condition observable. Its record binds source, time, method, and the current ART-01-01 fingerprint.
Contain the input contract: stop only the affected AI Search Setup path after observing “blocked or contradictory crawl directives”. Preserve its failed material and last verified state instead of erasing evidence or blindly repeating an external effect.
Recover and prove: apply the smallest authorized AI Search Setup correction, then have a distinct reviewer re-evaluate the complete accepted check set. Do not select one check merely because it shares this failure's list position. If any affected input contract check cannot run, its result remains NOT_TESTED.
2. Canonical links that point away from the intended page.
Detect for AI Search Setup: search implementation owner captures a direct readback or safe fixture that makes this input contract condition observable. Its record binds source, time, method, and the current ART-01-01 fingerprint.
Contain the input contract: stop only the affected AI Search Setup path after observing “canonical links that point away from the intended page”. Preserve its failed material and last verified state instead of erasing evidence or blindly repeating an external effect.
Recover and prove: apply the smallest authorized AI Search Setup correction, then have a distinct reviewer re-evaluate the complete accepted check set. Do not select one check merely because it shares this failure's list position. If any affected input contract check cannot run, its result remains NOT_TESTED.
3. Schema that parses but misdescribes the visible page.
Detect for AI Search Setup: content owner captures a direct readback or safe fixture that makes this input contract condition observable. Its record binds source, time, method, and the current ART-01-01 fingerprint.
Contain the input contract: stop only the affected AI Search Setup path after observing “schema that parses but misdescribes the visible page”. Preserve its failed material and last verified state instead of erasing evidence or blindly repeating an external effect.
Recover and prove: apply the smallest authorized AI Search Setup correction, then have a distinct reviewer re-evaluate the complete accepted check set. Do not select one check merely because it shares this failure's list position. If any affected input contract check cannot run, its result remains NOT_TESTED.
4. Orphaned pages absent from internal navigation.
Detect for AI Search Setup: release reviewer captures a direct readback or safe fixture that makes this input contract condition observable. Its record binds source, time, method, and the current ART-01-01 fingerprint.
Contain the input contract: stop only the affected AI Search Setup path after observing “orphaned pages absent from internal navigation”. Preserve its failed material and last verified state instead of erasing evidence or blindly repeating an external effect.
Recover and prove: apply the smallest authorized AI Search Setup correction, then have a distinct reviewer re-evaluate the complete accepted check set. Do not select one check merely because it shares this failure's list position. If any affected input contract check cannot run, its result remains NOT_TESTED.
5. Treating llms.txt as a substitute for useful content.
Detect for AI Search Setup: site owner captures a direct readback or safe fixture that makes this input contract condition observable. Its record binds source, time, method, and the current ART-01-01 fingerprint.
Contain the input contract: stop only the affected AI Search Setup path after observing “treating llms.txt as a substitute for useful content”. Preserve its failed material and last verified state instead of erasing evidence or blindly repeating an external effect.
Recover and prove: apply the smallest authorized AI Search Setup correction, then have a distinct reviewer re-evaluate the complete accepted check set. Do not select one check merely because it shares this failure's list position. If any affected input contract check cannot run, its result remains NOT_TESTED.
Ownership and handoff
| Role | Owned decision | Separation rule |
|---|---|---|
| site owner | owns the request boundary and confirms the intended consequence | May not approve evidence it produced when independent review is required |
| search implementation owner | owns the bounded implementation surface and action receipt | May not approve evidence it produced when independent review is required |
| content owner | owns source material, freshness, and the claim-to-evidence map | May not approve evidence it produced when independent review is required |
| release reviewer | owns release readiness, rollback, and destination verification | May not approve evidence it produced when independent review is required |
For this AI Search Setup input contract, the adjudication role is release reviewer. That role judges frozen acceptance evidence for search and AI-crawler discoverability without becoming the product owner, legal adviser, security authority, or buyer. Its handoff retains open gaps, failed evidence, changed hashes, and the next action permitted for ART-01-01.
Evidence and acceptance
Use these product-specific statements as candidate acceptance checks:
- Target URLs are fetchable without an unintended block.
- Canonical links resolve to the intended URLs.
- Sitemap entries match the canonical inventory.
- Structured data parses and agrees with visible content.
- The public llms.txt files expose the intended routes.
For every AI Search Setup input contract check, retain the tested object, environment or source, observation time, method, expected result, actual result, verifier identity, and artifact hash. In this ART-01-01 record, label a direct readback OBSERVED, a reproducible transformation COMPUTED, and an interpretation JUDGMENT; never merge those states into one confident claim.
The research packet observed 2 impressions across adjacent site queries such as “crawlable links” for the exact Search Console property https://sincllm.com/ during 2026-06-02/2026-08-30. Those observations help locate an existing audience vocabulary. They are not search-volume estimates, do not prove demand for this exact page, and do not predict clicks or rankings.
The product boundary remains controlling: The setup can make pages easier to discover and interpret, but it cannot guarantee rankings, citations, traffic, or inclusion in any model response.
Implementation checklist
- The input contract names the distinct reader job: Define the minimum input record and deterministic rejection rules before search and AI-crawler discoverability begins.
- The input boundary is explicit: domain access and a current inventory of the pages that should be discoverable.
- The intended deliverable is explicit: an indexing, schema, and llms.txt setup for the site.
- Every required acceptance check has current evidence or an honest NOT_TESTED status.
- At least one negative fixture covers blocked or contradictory crawl directives.
- The release reviewer is distinct from the artifact producer.
- Rollback or reopen conditions are written before consequential action.
- No ranking, traffic, conversion, compliance, certification, or buyer-outcome guarantee was added.
When this AI Search Setup input contract has a failed item, repair that named item and rerun its dependent checks. Keep the frozen threshold intact; the remaining checks cannot establish that the failed ART-01-01 condition probably holds.
Sources and claim boundaries
- sincLLM product catalog — used only for product capability and boundary.
- Google Search Central — used only for general procedure and control guidance.
- Google Search Central — used only for general procedure and control guidance.
- OpenAI documentation — used only for general procedure and control guidance.
For ART-01-01, the sincLLM catalog supplies the AI Search Setup product description. Its third-party references support only the general input contract procedure each source addresses. None proves a buyer-specific outcome from AI Search Setup or turns this page into a ranking, citation, or AI-answer guarantee.
Keep the AI Search Setup next step bounded
Review the catalog for this input contract, its required inputs, and its limits. Test any buyer-specific outcome from AI Search Setup in the buyer's environment instead of assuming it from the guide.
Explore the sincLLM product catalog