| 063698db3195efc4… | Pyrana has governance built in, including citations, typing, validation, human-in-the-loop controls, telemetry, and audit, and this built-in governance matches the RFP requirements line for line. | active | contextual | yes | |
| d4d07ade110b7ca3… | The recommended Pyrana on Fabric path adopts a pre-integrated, governed agentic platform that already runs on Azure and reads OneLake so that foundational platform integration is already in place. | active | contextual | yes | |
| 0d96d24d25cf40aa… | A direct Fabric build must subsequently engineer the governance demanded by the RFP, indicating that governance is not already fully integrated in that path. | active | contextual | yes | |
| 7444f5e2dcf4ebac… | Building directly on Fabric, identified as Path B, requires assembling Foundry Agent Service, Agent Framework, retrieval, vector store, a citation and provenance model, injection defenses, agent telemetry, and a custom frontend. | active | contextual | yes | |
| 928f3227f5f8b8d5… | The document states that the capability ceiling of a Fabric plus Foundry plus custom build approach is high, but the central issue is the cost of achieving and maintaining that capability under a requirement to deliver value quickly. | active | contextual | yes | |
| e3e2240ad3874a40… | Because Pyrana is configuration-driven, its handoff package consists of agent configs, knowledge sets, metric definitions, and audit trails as documentation-as-code that Summer Fridays IT can read, review, and own. | active | contextual | yes | |
| 7f39e88a684ba0e4… | On Pyrana, agent definitions, citations, controls, and tests are generated as governed artifacts rather than authored by hand so that governance documentation is produced systematically. | active | contextual | yes | |
| 6a22cdb0a892501b… | In the hybrid approach, the Fabric solution architecture, data model, and dashboards are delivered in Fabric, while the AI-agent inventory and governance approach are delivered on Pyrana. | active | contextual | yes | |
| a31e477842c56fcb… | The solution is considered production-ready only when IT has reviewed security and governance, UAT is complete, metrics and sources are documented, dashboards reconcile to Fabric, agents have documented controls and tests, and handoff is delivered. | active | contextual | yes | |
| e8caeb49d45af76d… | The RFP also requires testing, UAT, deployment, support, and handoff deliverables so that the implementation can be validated and transitioned into operations. | active | contextual | yes | |
| f1db58c2270ca0ce… | The RFP requires vendor deliverables including a project roadmap, Fabric solution architecture, data-source inventory and integration plan, data model and metrics and dashboard approach, AI-agent inventory and governance, security and access-control model, and data-validation and metric-definition approach. | active | contextual | yes | |
| 2ba2e0a010025b81… | Pyrana makes agents governed and testable by defining each agent in versioned configuration with documented purpose, sources, access, and limits, while telemetry and an evaluation harness support testing and monitoring. | active | contextual | yes | |
| 54aaa3b7bf9a384c… | Pyrana separates facts from recommendations by typing knowledge as fact, derived, or assumption and attaching confidence scores so that recommendations are never presented as validated facts. | active | contextual | yes | |
| 275f60f263bfe647… | Pyrana flags incomplete data through ambient gap detection that surfaces missing or stale knowledge before it reaches a recommendation so that downstream recommendations are informed by knowledge quality status. | active | contextual | yes | |
| d9a2b755426d8ae2… | Pyrana prevents invented claims by generating claims with the model but validating them with deterministic code against verbatim source text so that no claim is accepted without verification. | active | contextual | yes | |
| 3c2db49626dd4b0c… | Pyrana provides source traceability by making every Context Unit content-addressed with SHA-256 and attaching source, version, and approver metadata, while logging which units each agent answer used. | active | contextual | yes | |
| 95fc13e99d19727a… | This section states that Pyrana’s architecture is most directly aligned with the AI agent requirements because several listed requirements are already inherent platform behaviors rather than features to be added. | active | contextual | yes | |
| 348879947d0a8f18… | The RFP requires protection against prompt injection and requires strategic recommendations to be clearly distinguishable from validated business facts so that generated advice cannot be confused with verified information. | active | contextual | yes | |
| cc4f92155bbf2a41… | The RFP requires agents to identify when data is incomplete or unavailable and to avoid inventing metrics or conclusions so that users are not misled by unsupported outputs. | active | contextual | yes | |
| 6315b50f061b0bdb… | The RFP requires agents to use only approved sources, respect permissions, and provide source traceability so that outputs remain authorized and attributable. | active | contextual | yes | |
| fb25b4699e255577… | The RFP requires AI agents to be secure, governed, and limited to approved use cases, with each agent documented for purpose, sources, audience, access, limitations, testing, and monitoring so that operation remains controlled and auditable. | active | contextual | yes | |
| 1f327334acde9855… | The system must never expose raw sensitive data because Purview DLP and an output gateway normalize and validate every agent response while withholding raw confidential or customer-level data unless explicitly approved. | active | contextual | yes | |
| 4532399f8cfb3ee0… | Data validation and reconciliation are strong in Fabric, and agent claims are validated against source text before use so that outputs remain aligned with documented evidence. | active | contextual | yes | |
| 40481bc27a35ce0b… | Metric definitions and business logic are maintained in Fabric and Power BI so that dashboards reconcile to Fabric and agents and humans share one governed definition library. | active | contextual | yes | |
| 8e2f0bc2b1f6b19c… | AI agents must use only approved sources, respect permissions, provide source traceability, identify incomplete or unavailable data, avoid inventing metrics or conclusions, resist prompt injection, and clearly separate strategic recommendations from validated business facts so that outputs remain trustworthy and governed. | active | contextual | yes | |
| ff759c0b08c95ba8… | AI agents must be secure, governed, and limited to approved use cases, and each agent must document its purpose, sources, audience, access, limitations, testing, and monitoring so that agent deployment remains controlled and reviewable. | active | contextual | yes | |
| c38978f50c08a9b2… | To protect sensitive information, Purview DLP applies in Fabric-native, and Pyrana’s output gateway normalizes and validates every agent response while withholding raw confidential or customer-level data unless explicit approval is given so that unsafe disclosures are prevented. | active | contextual | yes | |
| 6ea9a56e9b81d037… | Data validation and reconciliation are strong in Fabric-native, while Pyrana reconciles dashboards to Fabric and validates agent claims against source text before use so that generated claims remain consistent with authoritative sources. | active | contextual | yes | |
| f2a798e46e7092a8… | Metric definitions and business logic documentation are maintained in Fabric and Power BI in the Fabric-native approach, while Pyrana captures them as governed Context Units so that agents and humans share a single definition library. | active | contextual | yes | |
| 54d0914dd758021a… | Source-to-AI traceability must be designed and built per agent in a Fabric-native approach, while Pyrana provides native traceability by logging the Context-Unit IDs used in each answer and linking each unit to the source document, version, approver, and verbatim quote so that AI outputs are auditable. | active | contextual | yes | |
| 56d8913fe54ed66b… | Source-to-dashboard traceability is strong in Fabric-native through Purview lineage, and that traceability is preserved in the Pyrana-on-Fabric approach for Power BI on Fabric so that dashboard lineage remains intact. | active | contextual | yes | |
| 09df279ff1ed11f5… | For the governed data model and metrics layer, Fabric-native uses a strong Fabric semantic model, while Pyrana consumes the same model and cites governed metric definitions instead of recomputing them so that agent outputs stay aligned with approved metrics. | active | contextual | yes | |
| 1995ccffbfe6c8bf… | The minimum security, governance, and data integrity requirements include Fabric solution architecture, workspace and environment structure, RBAC, data-source inventory, governed data models and metrics, validation and reconciliation, traceability, sensitive-data handling, production change management, and documentation handoff so that the full operating model is governed. | active | contextual | yes | |
| 010ffb47d002b1ad… | Agent governance, testing, and monitoring are a current gap in Fabric-native Assemble despite agent-level telemetry, while Pyrana provides native agent telemetry, an evaluation harness, audit logging, and human-in-the-loop approvals so that agent operations are governed more completely. | active | contextual | yes | |
| 59e07b315d4a26b2… | Production changes in Fabric-native are deployed through Fabric deployment pipelines, while Pyrana uses configuration-over-code in versioned YAML shipped through GitHub CI/CD so that new use cases can be introduced without platform changes. | active | contextual | yes | |
| 135caee2d91e90d6… | Sensitive data protection in Fabric-native uses Purview labels, DLP, and encryption, while Pyrana inherits those labels and adds private inference, no training on customer data, tenant isolation, and an output gateway that never exposes raw confidential data so that sensitive information remains protected in AI outputs. | active | contextual | yes | |
| 9979e31f53c55a1b… | Access control in the Fabric-native approach relies on Entra ID with Fabric and Purview RBAC, while Pyrana uses the same Entra ID and adds per-knowledge-set scoping and need-to-know controls down to the Context-Unit level so that access can be restricted more granularly. | active | contextual | yes | |
| d0932e366a9b1b84… | For ingestion, transformation, refresh, and security, the Fabric-native approach uses Fabric pipelines and Purview, while Pyrana leaves data handling unchanged and adds document extraction, validation, and storage with provenance and refresh monitoring so that document workflows are governed alongside data workflows. | active | contextual | yes | |
| be0f71a5976d1193… | In both approaches, OneLake remains the governed data store, while the recommended Pyrana-on-Fabric option adds a knowledge graph that stores only cited claims and provenance within the customer’s Azure tenant so that primary storage governance is unchanged. | active | contextual | yes | |
| bfb185bb4c853420… | IT must have visibility into and approval over data storage, ingestion, transformation, refresh, security, access, sensitive-data protection, production deployment changes, and AI-agent governance, testing, and monitoring so that platform and IT oversight requirements are satisfied. | active | contextual | yes | |
| 5fd809f5a210097a… | The requirement-by-requirement crosswalk presents every Summer Fridays RFP requirement category with the Fabric-native approach and the Pyrana-on-Fabric approach side by side so that a compliant and differentiated vendor response can be structured. | active | contextual | yes | |
| ea5f0eff1572c10f… | The build-versus-buy decision depends on whether the organization wants to build a governance layer itself and how long that effort would take, so that Section 6 can evaluate the tradeoff explicitly. | active | contextual | yes | |
| cc5b6f6955db3821… | AI agents must protect against prompt injection, and strategic recommendations must be clearly distinguishable from validated business facts so that generated advice cannot be confused with verified evidence. | active | contextual | yes | |
| 3e4690e4ffd5e917… | AI agents must identify when data is incomplete or unavailable and must avoid inventing metrics or conclusions so that users are not misled by unsupported outputs. | active | contextual | yes | |
| 3c462ad5f208a771… | AI agents must use only approved sources, respect permissions, and provide source traceability so that outputs are based on authorized and auditable information. | active | contextual | yes | |
| 4a44ecb94e203e2d… | Each AI agent must document its purpose, sources, audience, access, limitations, testing, and monitoring so that stakeholders can understand and govern the agent’s intended operation. | active | contextual | yes | |
| 423e4c60b8d3a2cf… | AI agents must be secure, governed, and limited to approved use cases so that agent deployment remains controlled and aligned with authorized business purposes. | active | contextual | yes | |
| 08e1e360e7d5aeee… | In the Pyrana-on-Fabric approach, the output gateway normalizes and validates every agent response and withholds raw confidential or customer-level data unless explicitly approved so that sensitive outputs are controlled before release. | active | contextual | yes | |
| 7fefb80ac6aaea3f… | In the Pyrana-on-Fabric approach, dashboards reconcile to Fabric and agent claims are validated against source text before use so that generated claims remain consistent with governed data and source evidence. | active | contextual | yes | |
| 1fc1f1886386bb62… | For data validation and reconciliation, the Fabric-native approach is described as strong so that native validation capabilities are considered robust. | active | contextual | yes | |