Intent detection: System design

Specifies prompts or rules, context, tools, state, structured outputs, retries, and system integration for intent detection. It fixes adoption boundaries and observable failure evidence while keeping environment-specific limits configurable.

Package status: reference context ready for human review. The contract and test scenarios are complete, but no claim is made that an adopting implementation has passed them.

Decision

For Intent detection, adopt a versioned orchestration policy and an explicit model-call boundary. Use one owned boundary, explicit state model, and versioned interfaces; keep policy decisions separate from adapters and side effects. This architecture block focuses specifically on system design. This package is a reference contract: the adopting team must choose numeric limits, providers, jurisdictions, and operational owners from evidence in its own environment.

Scope

• The Intent detection actors, inputs, outputs, states, policy or schema versions, and externally visible outcomes required for the system design focus of this architecture block.
• Primary and alternate interfaces, background work, caches, integrations, support paths, and evidence that can exercise the same boundary.
• Adoption-specific configuration, rollout, recovery, and verification responsibilities needed to apply the reference safely.

Outside this block

• Choosing a universal vendor, framework, jurisdiction, numeric threshold, retention period, or service-level objective for every adopter.
• Claiming that packaged scenarios have run against a downstream implementation or that this reference grants legal, security, or accessibility certification.

Contract

• Persist the prompt, tool, retrieval, model, and policy versions needed to reproduce or explain each material outcome.
• Validate structured output before any side effect; malformed or incomplete output is an error, not partial authority.
• Separate untrusted user or retrieved content from system policy and label each context item with origin and trust class.
• Bound tokens, tool calls, latency, and spend per request; crossing a bound follows a named fallback instead of continuing unbounded.
• Document component ownership and allowed dependency direction, including which data may cross each boundary.
• Make synchronous, asynchronous, and batch paths implement the same authority and state invariants.
• Version the intent taxonomy, definitions, examples, exclusions, multi-intent policy, unknown and out-of-scope states, language handling, confidence calibration, and routing consequences.
• Evaluate representative and adversarial utterances by slice, preserve the original input and model or rule evidence, ask or fall back below threshold, and never let inferred intent alone authorize a high-impact action.
• For the system design variant, enforce this declared boundary: specifies prompts or rules, context, tools, state, structured outputs, retries, and system integration for intent detection.

Implementation guidance

• Start with the domain state and interface contract, then attach adapters, storage, queues, and provider-specific code.
• Model Intent detection inputs, outputs, actors, states, invariants, side effects, and evidence before selecting framework or vendor details.
• Store the applicable Intent detection contract or policy version with material state so migrations, replay, and support decisions remain attributable.
• Introduce the path behind controlled rollout, compare expected and observed outcomes, and keep a tested recovery route until adoption evidence is complete.

Failure handling and safeguards

• Provider failure, unsafe output, or exhausted budget returns a bounded fallback and retains correlation-safe diagnostic evidence.
• Invalid or contradictory Intent detection state is rejected or quarantined; the implementation never guesses a value that broadens authority or duplicates an effect.
• Retries are bounded and use durable identity; after exhaustion, work reaches an inspectable terminal state with an accountable owner.

Verification and operations

• Exercise the contract through each adapter and prove that a failed dependency cannot bypass the authoritative boundary.
• Monitor Intent detection success, denial, validation failure, dependency failure, retry exhaustion, and recovery by contract version without sensitive payload dimensions.
• Re-run paired positive and negative fixtures after policy, schema, dependency, migration, or boundary changes and preserve the resulting evidence.

Adoption assumptions

• Names and values in the example are a concrete fixture, not universal defaults; adopters replace them through documented evidence and ownership.
• The adopting system can provide authenticated identity, durable operation or record versions, bounded telemetry, and a controlled path for change.

The executable-looking examples in this package are fixtures and acceptance contracts. Run the collection validator to check structure and metadata, then translate and execute the scenarios in the target repository before recording implementation evidence.

References

• NIST AI 600-1, Generative AI Profile (applies as of 2026-09-08)