Assume any text retrieved from a listing, email, PDF, vendor portal, or user can be hostile. Models process instructions and data through the same channel; your architecture must restore the boundary.
Treat this as an operating document, not a reading assignment. For Belle Realty or a property-management assistant, define the unit of work, owner, allowed failure modes, and the customer-visible consequence before choosing a model or framework. A system that gives plausible answers but cannot be measured, contained, or recovered is not production-ready.
Design stance
Start with a narrow contract. State what enters the boundary, what the component may read or change, what it must return, and when it must stop. Put the contract in version control alongside representative tenant, listing, maintenance, and leasing cases. Prefer a deterministic rule when one exists; use model judgment only where language or ambiguity genuinely adds value.
Operating loop
Instrument every request with a correlation ID, model and prompt version, retrieval version, tool calls, latency, token usage, policy decisions, and final outcome. Review a small, stratified sample of real traffic weekly. Track failures by property, workflow, language, channel, and tenant impact; aggregate averages hide the exact bad experience that produces churn.
Controls
Set an explicit threshold that changes behavior: block, require approval, degrade to search-only, route to a cheaper model, or page an owner. Test the threshold with synthetic failures and recent production examples. Do not make a dashboard metric a promise unless a named person can act on it within a defined window.
Production exercise
Apply this to a maintenance triage request: a resident reports a gas smell at 2 a.m. Document the safe path, the tool permissions, the human escalation, the message shown to the resident, and the evidence retained for review. Ship only after the behavior is repeatable in a rehearsal.
Exit criteria
The implementation has a written contract, measurable outcomes, a failure path, an accountable owner, and a rollback or containment action. If any is missing, it is still a prototype.
Never give an agent broad database access because it is convenient. Enforce tenant-scoped retrieval, allowlisted fields, short-lived credentials, redaction, egress controls, and auditable access decisions.
Treat this as an operating document, not a reading assignment. For Belle Realty or a property-management assistant, define the unit of work, owner, allowed failure modes, and the customer-visible consequence before choosing a model or framework. A system that gives plausible answers but cannot be measured, contained, or recovered is not production-ready.
Design stance
Start with a narrow contract. State what enters the boundary, what the component may read or change, what it must return, and when it must stop. Put the contract in version control alongside representative tenant, listing, maintenance, and leasing cases. Prefer a deterministic rule when one exists; use model judgment only where language or ambiguity genuinely adds value.
Operating loop
Instrument every request with a correlation ID, model and prompt version, retrieval version, tool calls, latency, token usage, policy decisions, and final outcome. Review a small, stratified sample of real traffic weekly. Track failures by property, workflow, language, channel, and tenant impact; aggregate averages hide the exact bad experience that produces churn.
Controls
Set an explicit threshold that changes behavior: block, require approval, degrade to search-only, route to a cheaper model, or page an owner. Test the threshold with synthetic failures and recent production examples. Do not make a dashboard metric a promise unless a named person can act on it within a defined window.
Production exercise
Apply this to a maintenance triage request: a resident reports a gas smell at 2 a.m. Document the safe path, the tool permissions, the human escalation, the message shown to the resident, and the evidence retained for review. Ship only after the behavior is repeatable in a rehearsal.
Exit criteria
The implementation has a written contract, measurable outcomes, a failure path, an accountable owner, and a rollback or containment action. If any is missing, it is still a prototype.
Jailbreak resistance is layered: policy-aware prompting, classifier or rule checks, constrained tools, output review, and human escalation for high-impact requests. Measure bypass rate against an evolving suite.
Treat this as an operating document, not a reading assignment. For Belle Realty or a property-management assistant, define the unit of work, owner, allowed failure modes, and the customer-visible consequence before choosing a model or framework. A system that gives plausible answers but cannot be measured, contained, or recovered is not production-ready.
Design stance
Start with a narrow contract. State what enters the boundary, what the component may read or change, what it must return, and when it must stop. Put the contract in version control alongside representative tenant, listing, maintenance, and leasing cases. Prefer a deterministic rule when one exists; use model judgment only where language or ambiguity genuinely adds value.
Operating loop
Instrument every request with a correlation ID, model and prompt version, retrieval version, tool calls, latency, token usage, policy decisions, and final outcome. Review a small, stratified sample of real traffic weekly. Track failures by property, workflow, language, channel, and tenant impact; aggregate averages hide the exact bad experience that produces churn.
Controls
Set an explicit threshold that changes behavior: block, require approval, degrade to search-only, route to a cheaper model, or page an owner. Test the threshold with synthetic failures and recent production examples. Do not make a dashboard metric a promise unless a named person can act on it within a defined window.
Production exercise
Apply this to a maintenance triage request: a resident reports a gas smell at 2 a.m. Document the safe path, the tool permissions, the human escalation, the message shown to the resident, and the evidence retained for review. Ship only after the behavior is repeatable in a rehearsal.
Exit criteria
The implementation has a written contract, measurable outcomes, a failure path, an accountable owner, and a rollback or containment action. If any is missing, it is still a prototype.
Separate untrusted content from system instructions, label it clearly, restrict tool authority, and validate tool arguments outside the model. Detection helps; permission design is the real control.
Treat this as an operating document, not a reading assignment. For Belle Realty or a property-management assistant, define the unit of work, owner, allowed failure modes, and the customer-visible consequence before choosing a model or framework. A system that gives plausible answers but cannot be measured, contained, or recovered is not production-ready.
Design stance
Start with a narrow contract. State what enters the boundary, what the component may read or change, what it must return, and when it must stop. Put the contract in version control alongside representative tenant, listing, maintenance, and leasing cases. Prefer a deterministic rule when one exists; use model judgment only where language or ambiguity genuinely adds value.
Operating loop
Instrument every request with a correlation ID, model and prompt version, retrieval version, tool calls, latency, token usage, policy decisions, and final outcome. Review a small, stratified sample of real traffic weekly. Track failures by property, workflow, language, channel, and tenant impact; aggregate averages hide the exact bad experience that produces churn.
Controls
Set an explicit threshold that changes behavior: block, require approval, degrade to search-only, route to a cheaper model, or page an owner. Test the threshold with synthetic failures and recent production examples. Do not make a dashboard metric a promise unless a named person can act on it within a defined window.
Production exercise
Apply this to a maintenance triage request: a resident reports a gas smell at 2 a.m. Document the safe path, the tool permissions, the human escalation, the message shown to the resident, and the evidence retained for review. Ship only after the behavior is repeatable in a rehearsal.
Exit criteria
The implementation has a written contract, measurable outcomes, a failure path, an accountable owner, and a rollback or containment action. If any is missing, it is still a prototype.
Run sanctioned attacks against prompts, documents, tools, identity boundaries, and operator workflows. Define scope, safe test data, stop conditions, evidence capture, severity scoring, remediation verification, and disclosure rules.
Treat this as an operating document, not a reading assignment. For Belle Realty or a property-management assistant, define the unit of work, owner, allowed failure modes, and the customer-visible consequence before choosing a model or framework. A system that gives plausible answers but cannot be measured, contained, or recovered is not production-ready.
Design stance
Start with a narrow contract. State what enters the boundary, what the component may read or change, what it must return, and when it must stop. Put the contract in version control alongside representative tenant, listing, maintenance, and leasing cases. Prefer a deterministic rule when one exists; use model judgment only where language or ambiguity genuinely adds value.
Operating loop
Instrument every request with a correlation ID, model and prompt version, retrieval version, tool calls, latency, token usage, policy decisions, and final outcome. Review a small, stratified sample of real traffic weekly. Track failures by property, workflow, language, channel, and tenant impact; aggregate averages hide the exact bad experience that produces churn.
Controls
Set an explicit threshold that changes behavior: block, require approval, degrade to search-only, route to a cheaper model, or page an owner. Test the threshold with synthetic failures and recent production examples. Do not make a dashboard metric a promise unless a named person can act on it within a defined window.
Production exercise
Apply this to a maintenance triage request: a resident reports a gas smell at 2 a.m. Document the safe path, the tool permissions, the human escalation, the message shown to the resident, and the evidence retained for review. Ship only after the behavior is repeatable in a rehearsal.
Exit criteria
The implementation has a written contract, measurable outcomes, a failure path, an accountable owner, and a rollback or containment action. If any is missing, it is still a prototype.