Principal-led architecture for critical systems

Trust Center

Security, data, and AI operating boundaries

Public practices, limitations, and document-request paths are visible without implying certification, audit, authorization, or universal compliance.

Trust CenterPublic operating boundaries
Purpose
Review public security, data, AI-authority, accessibility, continuity, and document-request practices.
Boundary
Certification, audit, authorization, insurance, procurement status, and universal compliance are not implied.

Public operating approach

Security overview

The public site remains static-first for public content, uses local assets and local search, and does not configure external analytics or a third-party lead platform. An optional first-party endpoint stores only consented, public-safe inquiry data in a separately installed private application; sensitive technical material remains prohibited. Client security requirements are qualified and scoped before access.

  • Least authority and role separation
  • Secure credential handling through approved client channels
  • Version control, review gates, logging, and rollback where scoped
  • Environment and data-boundary separation
  • Data minimization and dependency review
  • No secrets committed to public source or client-side code
  • Private lead configuration, keys, application logic, and database access remain outside the public document root
  • Lead content is encrypted at rest and can be cryptographically redacted when the private application is installed

Public classification guide

Data-handling boundaries

Public channels are limited to public-safe qualification. Confidential or regulated evidence moves only after responsibility, scope, contractual terms, and an approved transfer method are clear.

  • Public-safe: organization, role, public notice links, high-level system context, timing, and budget range
  • Confidential business information: accepted only through an approved private channel after qualification
  • Regulated or sensitive information: requires explicit contractual and technical handling decisions
  • Prohibited through public forms and email: credentials, private source code, customer records, PHI, financial account data, classified, CUI, export-controlled, source-selection-sensitive, or privileged material

Operating boundary

AI use and human authority

AI output is treated as proposed work. Consequential action requires explicit human authority, and generated claims require source support or an unknown state.

  • Client data is not placed in unapproved tools
  • Data, retention, and model-use boundaries are defined for the engagement
  • Human review remains explicit for consequential work
  • AI use is disclosed when materially relevant
  • Unknowns are not converted into certainty
  • Approval and execution are separated where risk warrants

Use the AI System Intake & Release Decision Record

Engineering practice

Secure software practices

Engagements may use secure-development concepts appropriate to the system, including least authority, reviewable change, dependency management, versioned evidence, test gates, rollback, and explicit operational ownership.

  • Source and change control
  • Dependency and configuration review
  • Secrets excluded from source
  • Review and release gates
  • Logging, observability, and rollback design
  • Client-owned artifact export and handoff

Delivery principle

Business continuity and exit

The delivery model favors client-owned artifacts, exportable formats, documented dependencies, knowledge transfer, and a planned handoff rather than manufactured lock-in.

  • Client retains client-specific deliverables
  • Pre-existing methods and templates remain provider IP
  • Dependencies and external services are documented
  • Decision and risk records are exportable
  • Handoff and exit expectations are named in scope
  • Additional personnel are disclosed and scoped

Buyer-operable assurance path

Trust document catalog

Start with public material, then request only the engagement-specific evidence needed for the decision. Access status is explicit so a request does not imply that an audit, certificate, report, or prebuilt response exists.

ItemAccessHelps answerBoundaryAction
Security overviewPublicWhat public security and delivery practices are currently described?A public operating overview, not an audit, certification, penetration test, or security guarantee.Download PDF
Public data, AI, privacy, accessibility, and continuity boundariesPublicWhat can be reviewed before sharing confidential or regulated information?Public statements describe current practices and limits; they do not establish universal compliance.Review boundaries
Security or customer-assurance questionnaire responseRequestableCan LongTermCapabilities answer the questions relevant to a specific procurement or engagement?Responses are scoped to the actual service, tools, data, responsibilities, and available evidence; no blanket attestation is implied.Prepare request
Engagement data-flow and handling boundaryRequestableWhich data, environments, tools, people, transfers, retention rules, and exclusions apply to the proposed work?The useful answer depends on a bounded engagement and client requirements; sensitive detail moves through an approved channel.Prepare request
AI use and human-authority walkthroughRequestableWhere may AI assist, who reviews it, and who retains authority over consequential action?The walkthrough describes the proposed engagement and does not certify the buyer's broader AI program.Prepare request
Secure delivery, continuity, and exit responseRequestableHow will change, access, evidence, handoff, continuity, and exit be handled for the workstream?Controls and artifacts are selected for the scoped work; no 24/7 operations or universal recovery guarantee is implied.Prepare request
Engagement-specific tool, dependency, and subprocessor disclosureAfter scopeWhich third-party tools or services would actually be used for the proposed engagement?No universal list is asserted before scope. The applicable set is disclosed and approved for the engagement.Prepare request
Independent assurance reports and certificationsNot assertedIs an independent audit, SOC report, ISO certificate, penetration-test report, or comparable assurance artifact publicly claimed?No such artifact is publicly asserted as currently held. Unknown or unavailable evidence is not presented as complete.No public artifact claimed

Open the machine-readable catalog

Disclosure status

DisclosureStatusPublic statementReviewed
Public-site architecturePublicStatic-first public content, local assets, local search, optional first-party public-safe lead intake, and no configured advertising pixels, external analytics, hosted search, external fonts, third-party lead platform, public chatbot, or live MCP transport.2026-07-24
Data-handling boundariesPublicPublic forms and public email are limited to public-safe qualification. Confidential or regulated evidence requires an approved channel after scope and responsibility are clear.2026-07-24
AI use and human authorityPublicAI output is proposed work. Consequential actions require explicit authority, source support, review, and blocked-action criteria appropriate to the engagement.2026-07-24
Secure software practicesConditionalEngagement controls are selected for the system and may include least authority, version control, dependency review, test and release gates, rollback, environment separation, and secret exclusion.2026-07-24
Framework alignmentPublicThe practice can map technical work to public frameworks for readiness and evidence discussions. Certification, attestation, approval, and universal compliance are not asserted.2026-07-24
Procurement and insurance statusPrivate on requestVerified entity, registration, insurance, and legal documents are shared through an approved procurement channel when relevant. Unverified values are not displayed.2026-07-24

Document request process

Use the catalog to identify the smallest relevant request. Public documents are available directly; engagement-specific material is qualified before private exchange.

  • Legal entity details
  • W-9
  • Insurance certificates
  • Registration records
  • Security materials
  • Standard agreements
  • Other opportunity-specific vendor documents

Next action

Ask for the evidence appropriate to the decision.

A public page is not a substitute for due diligence, and private materials are not sent before fit and handling are established.

Private local search

Find a service, capability, evidence record, resource, or insight

Press / to open search when focus is not in a form field.

Search runs locally against the public site index.