| e7a89d2f052c94b8… | Within Phase 2, all remaining client folders must be extracted into their respective sub-sets after the named client folders so that residual client knowledge is classified by destination subset. | active | contextual | yes | |
| d9355de713f64f36… | Within Phase 2, the Tolmar folder must be extracted into CLIENTS:TOLMAR after JNJ so that Tolmar knowledge is ingested before the remaining client folders. | active | contextual | yes | |
| f13b0244fe47cfb0… | Within Phase 2, the JNJ folder must be extracted into CLIENTS:JNJ after Patrick so that JNJ client knowledge is loaded as the next named client priority. | active | contextual | yes | |
| cc808f73a3dc6187… | Within Phase 2, the Patrick folder must be extracted first because it is the highest-activity client, and its content must map to CLIENTS:PATRICK and PROJECTS:PATRICK_* so that the most active client knowledge is prioritized. | active | contextual | yes | |
| a94d2890cc920c5b… | Phase 2 of the extraction priority order must ingest client knowledge during weeks 3 through 4, starting with the highest-activity client and then named client folders before remaining client folders. | active | contextual | yes | |
| 02bb4d9a73227520… | Within Phase 1, core positioning materials must be extracted into SALES_POSITIONING as Rank 0 CXUs so that foundational sales positioning is prioritized at the highest rank. | active | contextual | yes | |
| f46a18447ee782ed… | Within Phase 1, company policies must be extracted into COMPANY_POLICIES as the third priority so that policy knowledge is added during the foundation stage. | active | contextual | yes | |
| b0768411bd508528… | Within Phase 1, contract templates must be extracted into CONTRACT_STANDARDS as the second priority so that standard contractual knowledge follows platform ingestion. | active | contextual | yes | |
| db157d0a8df640d6… | Within Phase 1, platform docs from GitHub repos must be extracted into PYRANA_PLATFORM as the first priority so that platform knowledge is loaded before other domains. | active | contextual | yes | |
| fd8c4311c4029831… | Phase 1 of the extraction priority order must ingest platform docs, contract templates, company policies, and core positioning materials during weeks 1 through 2 so that foundational knowledge is established first. | active | contextual | yes | |
| db345a34e9a594d2… | The candidate sets for LLM-based set assignment must include UNIVERSAL, PYRANA_PLATFORM, CONTRACT_STANDARDS, COMPANY_POLICIES, CLIENTS, PROJECTS, SALES_POSITIONING, SECURITY_COMPLIANCE, MARKETING_CONTENT, IP_PATENTS, FINANCIAL_OPS, and TEAM_NETWORK so that classification is constrained to approved destinations. | active | contextual | yes | |
| 85ca6a00fabaee2f… | The zeroth-internal profile must assign sets using an LLM-based classifier strategy over a defined candidate set list so that extracted content is routed to approved internal sets. | active | contextual | yes | |
| a0f88b7ea7ec0faf… | The zeroth-internal profile must deduplicate extracted content using a similarity threshold of 0.92 and prefer higher-confidence results so that near-duplicate units are resolved consistently. | active | contextual | yes | |
| c3ebb426ed9ea0da… | The zeroth-internal profile must use semantic chunking with a maximum chunk size of 2000 and overlap of 200 so that extraction operates on semantically grouped text windows. | active | contextual | yes | |
| 7264ba39799bc7bd… | The zeroth-internal extraction profile must include the facts, definitions, procedures, requirements, and specifications lenses so that internal extraction covers all configured knowledge categories. | active | contextual | yes | |
| ec1ed03b9ba92174… | A custom extraction profile named zeroth-internal must be created alongside the existing general and pharma-manufacturing profiles so that internal knowledge extraction uses a dedicated configuration. | active | contextual | yes | |
| e579472a6d4056a1… | The specifications lens must extract technical specifications for PYRANA_PLATFORM and PROJECTS so that technical implementation details are captured in those sets. | active | contextual | yes | |
| 5cb4426527b9af1c… | The requirements lens must extract must, shall, and required statements for CONTRACT_STANDARDS, COMPANY_POLICIES, and SECURITY_COMPLIANCE so that obligation-oriented knowledge is preserved. | active | contextual | yes | |
| 68fdb68b3774bb3c… | The procedures lens must extract step-by-step processes for COMPANY_POLICIES and CONTRACT_STANDARDS so that procedural knowledge is captured from policy and contract sources. | active | contextual | yes | |
| f47f552f3fe3af00… | The definitions lens must extract term definitions and concepts for PYRANA_PLATFORM and CONTRACT_STANDARDS so that terminology and conceptual knowledge are structured for those sets. | active | contextual | yes | |
| 58b2d082d90fe92d… | The facts lens must extract factual assertions across all sets during internal extraction so that general factual knowledge is captured universally. | active | contextual | yes | |
| 48ea1f80a96d9ee4… | The context engine must run multiple extraction lenses in parallel during the chunking phase for internal deployment so that different knowledge types are captured across designated primary sets. | active | contextual | yes | |
| 34400ac4e22c1d6b… | The source document inventory defines nine source categories with approximate file counts, primary formats, and target sets so that extraction can be planned against specific repositories and business domains. | active | contextual | yes | |
| ce7d8cea9a2f1c53… | The retrieval configuration defines a known query pattern where the question 'What is our security policy on X?' should map to a template for COMPANY_POLICIES lookup so that security policy questions are routed to policy retrieval. | active | contextual | yes | |
| a59c020bbf49300b… | The retrieval configuration defines a known query pattern where the question 'What platform components does Patrick use?' should map to a template for IMPLEMENTS relationship traversal so that client-platform usage questions are answered through graph relationship navigation. | active | contextual | yes | |
| 973e4127437937b0… | The retrieval configuration defines a known query pattern where the question 'What are the standard payment terms?' should map to a template for CONTRACT_STANDARDS requirements so that payment-term questions are routed to contractual requirement retrieval. | active | contextual | yes | |
| 4449167e49585790… | Strategy 1 is Template Cypher and should receive high weight because most internal queries will match known patterns so that common internal questions can be answered through predefined graph query templates. | active | contextual | yes | |
| 66c797c7a2dc8b15… | The retrieval configuration states that the 3-strategy retrieval pipeline should be configured for internal knowledge patterns so that retrieval behavior aligns with the kinds of questions expected in internal use. | active | contextual | yes | |
| 84a6d0bbfdfbbaba… | Phase 4 of the extraction priority order is the Ongoing Ingestion phase and includes continuous Granola meeting notes to CLIENTS and PROJECTS, automatic classification and extraction of new Drive documents, and daily research pipeline findings to PYRANA_PLATFORM as new research insights so that ingestion continues after the initial rollout. | active | contextual | yes | |
| fc53fcd4f6f9ab50… | Phase 3 of the extraction priority order is the Operational Knowledge phase scheduled for Week 5-6 and includes remaining sales materials, security and compliance documents, financial documents, patent documents, and remaining Shared Drive content mapped to their respective sets so that operational knowledge is ingested after client knowledge. | active | contextual | yes | |
| 2d8a0e6cbf6c0716… | Phase 2 of the extraction priority order is the Client Knowledge phase scheduled for Week 3-4 and includes the Patrick folder to CLIENTS:PATRICK and PROJECTS:PATRICK_*, the JNJ folder to CLIENTS:JNJ, the Tolmar folder to CLIENTS:TOLMAR, and remaining client folders to their respective subsets so that client-specific knowledge is ingested after the foundation phase. | active | contextual | yes | |
| 2d70713272af2717… | Phase 1 of the extraction priority order is the Foundation phase scheduled for Week 1-2 and includes platform docs to PYRANA_PLATFORM, contract templates to CONTRACT_STANDARDS, company policies to COMPANY_POLICIES, and core positioning materials to SALES_POSITIONING as Rank 0 CXUs so that foundational knowledge is ingested first. | active | contextual | yes | |
| 6c1e9fc40502f3b4… | The zeroth-internal extraction profile must use classifier-based set assignment, defined as LLM-based set assignment, with candidate sets including UNIVERSAL, PYRANA_PLATFORM, CONTRACT_STANDARDS, COMPANY_POLICIES, CLIENTS, PROJECTS, SALES_POSITIONING, SECURITY_COMPLIANCE, MARKETING_CONTENT, IP_PATENTS, FINANCIAL_OPS, and TEAM_NETWORK so that extracted content can be routed to approved internal sets. | active | contextual | yes | |
| 7bac1b4784e82a1a… | The zeroth-internal extraction profile must configure deduplication with a similarity threshold of 0.92 and a preference for higher confidence so that near-duplicate extractions are resolved in favor of stronger evidence. | active | contextual | yes | |
| 27d6874a128c2a94… | The zeroth-internal extraction profile must use a semantic chunking strategy with a maximum chunk size of 2000 and an overlap of 200 so that content is segmented with semantic boundaries and controlled context carryover. | active | contextual | yes | |
| 88a35b570ddcf56c… | The zeroth-internal extraction profile must include the facts, definitions, procedures, requirements, and specifications lenses so that internal extraction covers factual, conceptual, procedural, normative, and technical knowledge types. | active | contextual | yes | |
| 93223871d13175db… | A custom extraction profile named zeroth-internal must be created alongside the existing general and pharma-manufacturing profiles so that internal deployment uses a dedicated extraction configuration. | active | contextual | yes | |
| 5e71a3e7123a8463… | The specifications lens is defined as an extraction lens whose purpose is to extract technical specifications from the PYRANA_PLATFORM and PROJECTS sets so that technical implementation details are captured from platform and project materials. | active | contextual | yes | |
| 8473a2d89d04865b… | The requirements lens is defined as an extraction lens whose purpose is to extract must, shall, and required statements from the CONTRACT_STANDARDS, COMPANY_POLICIES, and SECURITY_COMPLIANCE sets so that normative obligations are captured from governance documents. | active | contextual | yes | |
| 2c8ccb04415a1468… | The procedures lens is defined as an extraction lens whose purpose is to extract step-by-step processes from the COMPANY_POLICIES and CONTRACT_STANDARDS sets so that procedural knowledge is captured from policy and contract materials. | active | contextual | yes | |
| f7e7dc6546fc05aa… | The definitions lens is defined as an extraction lens whose purpose is to extract term definitions and concepts from the PYRANA_PLATFORM and CONTRACT_STANDARDS sets so that conceptual and terminology knowledge is captured from those domains. | active | contextual | yes | |
| 7a4ef9bac12efee1… | The facts lens is defined as an extraction lens whose purpose is to extract factual assertions from all sets so that general factual knowledge is captured across the entire corpus. | active | contextual | yes | |
| 64985c2f466f951d… | The context engine supports multiple extraction lenses that run in parallel during the chunking phase for internal deployment so that different knowledge types can be extracted simultaneously from the same content. | active | contextual | yes | |
| bb25949c28e31504… | The source document inventory defines financial documents as approximately 8 Drive files in .xlsx and .pdf formats that should be assigned to the FINANCIAL_OPS target set so that financial operations knowledge is extracted from spreadsheets and documents. | active | contextual | yes | |
| 81cb9388af15ce4c… | The source document inventory defines meeting notes from Granola as an ongoing transcript source that should be assigned to the CLIENTS and PROJECTS target sets so that continuously generated meeting knowledge is captured. | active | contextual | yes | |
| 9f25d3ad477ff08b… | The source document inventory defines patent documents as a Shared Drive source with a file count still to be determined, using various formats, that should be assigned to the IP_PATENTS target set so that intellectual property records can be ingested. | active | contextual | yes | |
| 62f6bf9c033a4a9e… | The source document inventory defines marketing content as a Shared Drive source with a file count still to be determined, using various formats, that should be assigned to the MARKETING_CONTENT target set so that future marketing assets can be incorporated. | active | contextual | yes | |
| 80f1a1de248154b0… | The source document inventory defines sales materials as approximately 15 Drive files in .pptx, .pdf, and .docx formats that should be assigned to the SALES_POSITIONING target set so that sales positioning knowledge is ingested from presentation and document assets. | active | contextual | yes | |
| 1ba5f1805441c1f2… | The source document inventory defines company policies as approximately 6 Drive files in .pdf and .docx formats that should be assigned to the COMPANY_POLICIES and SECURITY_COMPLIANCE target sets so that policy and compliance knowledge is extracted together. | active | contextual | yes | |
| 83e1362414c285b8… | The source document inventory defines client folders as more than 80 Drive files in .docx, .pptx, .xlsx, and .pdf formats that should be assigned to the CLIENTS and PROJECTS target sets so that client and project knowledge is captured from mixed-format folders. | active | contextual | yes | |