| da4a71e46b391a98… | Standard Zeroth Agents MSA payment terms require payment Net 30 from the invoice date, with late payments accruing interest at 1.5% per month or the maximum rate permitted by law. | active | contextual | yes | |
| fa5f2e6f37ebda27… | The native tool-calling loop in PyranaHarness operates by having the LLM emit tool calls, executing those tools, observing results, and repeating the cycle until completion. | active | contextual | yes | |
| 26e40b250b84aed3… | The PyranaHarness implements native LLM tool calling through LiteLLM, executing a Turn 0 planning step followed by iterative tool-calling loops until it finalizes an AgentOutputEnvelope. | active | contextual | yes | |
| 4a09bba832800957… | Confidence scores below 0.40 indicate speculative knowledge that should be used only for gap-filling, so that draft ideas and unvalidated hypotheses are not treated as reliable evidence. | active | contextual | yes | |
| 3cd653437d021285… | Confidence scores from 0.40 to 0.59 indicate low-confidence knowledge that should be marked as an assumption when used, so that uncertain market estimates are distinguished from validated facts. | active | contextual | yes | |
| d1aa30313ba37285… | The PYRANA_PLATFORM example classifies the PyranaHarness claim as knowledge_type axiom, claim_type specification, confidence 0.98, with keywords PyranaHarness, orchestration, tool-calling, LiteLLM, and AgentOutputEnvelope. | active | contextual | yes | |
| 5831043b10bc52f4… | The PYRANA_PLATFORM example supports its claim with quotes stating that PyranaHarness implements native tool calling via LiteLLM, creates an initial Turn 0 plan, repeats a native tool-calling loop, and finalizes by building and persisting an AgentOutputEnvelope. | active | contextual | yes | |
| 4ec6bc3f6fdd9a0c… | In the provided PYRANA_PLATFORM example, the claim states that PyranaHarness implements native LLM tool calling via LiteLLM, executes a Turn 0 plan, iterates through tool-calling loops, and finalizes an AgentOutputEnvelope. | active | contextual | yes | |
| afd62b99f109f936… | A confidence score below 0.40 indicates speculative content intended only for gap-filling, including draft ideas and unvalidated hypotheses. | active | contextual | yes | |
| 5bf6ecf798a43a0c… | A confidence score from 0.40 to 0.59 indicates low confidence and should be marked as an assumption, such as the market estimate that the TAM for agentic platforms is $50B by 2028. | active | contextual | yes | |
| a9dbeadc845409e2… | A confidence score from 0.60 to 0.79 indicates moderate confidence and possible need for updating, such as the estimate that standard implementation takes 6 to 8 weeks based on limited projects. | active | contextual | yes | |
| 4aa1589bda2b5a3b… | A confidence score from 0.80 to 0.94 indicates high confidence from a reliable source, such as the statement that Patrick's financial analysis covers more than 50 subsidiaries from an SOW. | active | contextual | yes | |
| 625a2632c5cbf8ca… | A confidence score from 0.95 to 1.0 indicates a verified fact from an authoritative source, such as the statement that PYRANA uses PostgreSQL, Neo4j, and Qdrant from CLAUDE.md. | active | contextual | yes | |
| 263847592fd524f1… | The prescribed knowledge type should be used for standards, policies, and approved language, and it requires regulated explicit approval so that controlled organizational guidance is not accepted automatically. | active | contextual | yes | |
| 62c9bbcea1f4d6ab… | The derived knowledge type should be used for conclusions drawn from data or experience, and it follows the standard approval workflow so that inferred operational insights are categorized separately from facts and policies. | active | contextual | yes | |
| 05c4e432a3de55ed… | The axiom knowledge type should be used for stable facts that rarely change, and it follows the standard auto-approve workflow so that durable foundational information can be accepted with minimal review. | active | contextual | yes | |
| 3cb0620f44292cc3… | Every CXU stored in the graph must track lifecycle using the sequence DRAFT to ACTIVE to INACTIVE to VERSIONED so that unit state can be managed over time. | active | contextual | yes | |
| 64bb536361f76ff9… | Every CXU stored in the graph must include a confidence field as a float from 0.0 to 1.0 so that the certainty of the extracted knowledge is quantitatively represented. | active | contextual | yes | |
| 34eb9662930e37fd… | Every CXU stored in the graph must include keywords as a list of strings used for full-text search and filtering so that units can be efficiently retrieved and narrowed. | active | contextual | yes | |
| 0747c6bba9e15f2e… | Every CXU stored in the graph must include parent_assumption_ids as a list of strings representing chains of dependent assumptions so that assumption lineage can be tracked. | active | contextual | yes | |
| 89ee9c05b0acc4f4… | Every CXU stored in the graph must include an is_assumption boolean that is set to true for projections and estimates so that assumption-bearing knowledge is explicitly marked. | active | contextual | yes | |
| 823f6ce2c9eaa5cf… | Every CXU stored in the graph must assign claim_type from the set definition, requirement, procedure, specification, or relationship so that the form of the claim is explicitly classified. | active | contextual | yes | |
| 1a09b3d954a29ef0… | Every CXU stored in the graph must assign knowledge_type from the set axiom, derived, or prescribed so that each unit is categorized by its epistemic status. | active | contextual | yes | |
| d208c1efd1056db9… | Every CXU stored in the graph must include supporting_quotes as a list of 3 to 6 verbatim source quotes so that the claim is grounded in explicit source evidence. | active | contextual | yes | |
| d91bb70f3a0ee270… | Every CXU stored in the graph must contain a claim field as a standalone assertion with a length of 15 to 70 words so that each unit remains self-contained and atomic. | active | contextual | yes | |
| f4d8e85a57f592d9… | Every CXU stored in the graph must include a content-addressable cxu_id formed as "1220" plus SHA-256(claim), producing a 68-character identifier so that each unit is deterministically keyed by its claim content. | active | contextual | yes | |
| 18198b05bea890f6… | To draft an NDA for a new biotech prospect, the graph traversal should begin with CONTRACT_STANDARDS:NDA, follow CUSTOMIZED_FROM to CLIENTS:CELLDEX, and then follow CONSTRAINS to COMPANY_POLICIES so that prior biotech customizations and required policy constraints are included. | active | contextual | yes | |
| 143e58813c335300… | To determine what platform capabilities the Patrick financial analysis project uses, the graph traversal should start at PROJECTS:PATRICK_FINANCIAL_ANALYSIS and follow IMPLEMENTS to PYRANA_PLATFORM:CORE and PYRANA_PLATFORM:CODE_INTERPRETER. | active | contextual | yes | |
| eb03793b72797c9f… | To prepare a proposal for a new pharma client, the graph traversal should start from SALES_POSITIONING:VALUE_PROPS, follow POSITIONS_FOR to CLIENTS:LIFE_SCIENCES, then SCOPED_BY to PROJECTS:TOLMAR_* and PROJECTS:JNJ_*, and finally DELIVERED_UNDER to CONTRACT_STANDARDS:SOW so that relevant pharma project history and SOW templates are retrieved. | active | contextual | yes | |
| f54e663dc684f123… | Cross-set query patterns enable multi-hop traversal across relationship types so that users can answer compound business questions by linking knowledge from multiple graph sets. | active | contextual | yes | |
| 6d0d7b3c276b732d… | The example platform CXU uses a confidence score of 0.98 and keywords including PyranaHarness, orchestration, tool-calling, LiteLLM, and AgentOutputEnvelope so that the unit is both highly trusted and searchable by core technical terms. | active | contextual | yes | |
| 1a77d170b89dfdea… | The example platform CXU states that PyranaHarness implements native LLM tool calling via LiteLLM by executing a Turn 0 plan followed by iterative tool-calling loops until an AgentOutputEnvelope is finalized. | active | contextual | yes | |
| 710fbfdccf9a2d84… | The example PYRANA_PLATFORM CXU is classified as axiom knowledge and specification claim type, showing that platform implementation facts can be represented as stable technical specifications. | active | contextual | yes | |
| 97a57d92b341920c… | Confidence scores below 0.40 indicate speculative content intended only for gap-filling, including draft ideas and unvalidated hypotheses so that highly uncertain material is clearly limited in use. | active | contextual | yes | |
| 0fa8fe0d7754437e… | Confidence scores from 0.40 to 0.59 indicate low confidence and should be marked as assumptions, such as a market estimate that the TAM for agentic platforms will reach $50B by 2028. | active | contextual | yes | |
| 0b703530e85c3412… | Confidence scores from 0.60 to 0.79 indicate moderate confidence and possible need for updating, such as an estimate that standard implementation takes 6 to 8 weeks based on limited projects. | active | contextual | yes | |
| c8783c8379e40cc8… | Confidence scores from 0.80 to 0.94 indicate high confidence from a reliable source, such as a statement that Patrick's financial analysis covers more than 50 subsidiaries when sourced from an SOW. | active | contextual | yes | |
| 19f13b58e63f6cca… | Confidence scores from 0.95 to 1.0 indicate a verified fact from an authoritative source, such as the documented statement that PYRANA uses PostgreSQL, Neo4j, and Qdrant. | active | contextual | yes | |
| d70a9f07a0572efb… | Prescribed knowledge should be used for standards, policies, and approved language, and it requires regulated explicit approval so that controlled organizational content is reviewed before use. | active | contextual | yes | |
| e1c4032aef76c0f8… | Derived knowledge should be used for conclusions drawn from data or experience, and it follows the standard approval workflow so that inferred operational insights are distinguished from raw facts and policies. | active | contextual | yes | |
| b84f6489af3a0f66… | Axiom knowledge should be used for stable facts that rarely change, and it follows the standard auto-approve workflow so that enduring factual information can be added with minimal review overhead. | active | contextual | yes | |
| 64d8860bf9dfcf6a… | The lifecycle field in a CXU is defined to progress through DRAFT, ACTIVE, INACTIVE, and VERSIONED states so that each unit’s status is tracked over time. | active | contextual | yes | |
| 2e2fdde52c31f1c0… | The confidence field in a CXU is defined as a float from 0.0 to 1.0 so that each unit carries an explicit quantitative estimate of certainty. | active | contextual | yes | |
| cfcef0cd8aef2676… | The keywords field in a CXU is defined as a list of terms used for full-text search and filtering so that units can be efficiently retrieved and narrowed by topic. | active | contextual | yes | |
| a03c8e21aa6df7c5… | The parent_assumption_ids field in a CXU is defined as a list of strings representing chains of dependent assumptions so that assumption lineage can be traced across related units. | active | contextual | yes | |
| fd27beceebf9d0ff… | The is_assumption field in a CXU is defined as a boolean that is true for projections and estimates so that uncertain forward-looking content is explicitly marked. | active | contextual | yes | |
| 86c7e879381e373f… | The claim_type field in a CXU is defined to use the values definition, requirement, procedure, specification, or relationship so that the form of the claim is explicitly classified. | active | contextual | yes | |
| 908fa83770fa66a0… | The knowledge_type field in a CXU is defined to use the values axiom, derived, or prescribed so that each unit is categorized by the nature of the knowledge it contains. | active | contextual | yes | |
| 59f33d0c05edfd72… | The supporting_quotes field in a CXU is defined as a list of 3 to 6 verbatim source quotes so that each claim retains direct textual evidence from the original material. | active | contextual | yes | |
| fe9e87e61ff80bb9… | The claim field in a CXU is defined as a standalone assertion between 15 and 70 words so that each extracted unit can be understood independently of its source document. | active | contextual | yes | |