Documentation

Reference

Examples

Three full packages compiled live from real source text. Same compiler, three domains. Copy, download or paste into your own pipeline.

Packages

Learning science

ckf_demo_1778889423741

8 entities · 6 concepts · 0 principles
markdown
# CKF — KNOWLEDGE CONTEXT PACKAGE

package_id: ckf_demo_1778889423741
protocol_version: ckf-0.1
source_type: article
source_title: Untitled source
source_author: Unknown
domain: education / learning science
subdomains: [students, rereading, memory, sleep]
language: en
created_at: 2026-05-15T23:57:03.741Z
compression_level: standard
human_readability: 0.7
ai_utility_score: 0.81

---

## 1. CORE INTENT

```yaml
core_intent:
  primary_purpose: Capture and structure the knowledge expressed in the source.
  intended_user: Developers, researchers and agents consuming structured knowledge.
  intended_agent_use: Retrieval, reasoning, tutoring, decision support.
  transformation_goal: Convert prose into structured, agent-usable cognition.
  key_value: Portable, traceable, reusable knowledge package.
```

## 2. DOMAIN MAP

```yaml
domain_map:
  main_domain: education / learning science
  subdomains:
    - name: students
      relevance: 1
      related_concepts:
        - students
    - name: rereading
      relevance: 0.85
      related_concepts:
        - rereading
    - name: memory
      relevance: 0.7
      related_concepts:
        - memory
    - name: sleep
      relevance: 0.55
      related_concepts:
        - sleep
  adjacent_domains: []
  excluded_domains: []
```

## 3. ENTITY GRAPH

```yaml
entities:
  - id: ENT_001
    name: Learning
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities: []
    source_basis: explicit
  - id: ENT_002
    name: Spaced
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_003
    name: Sleep
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_004
    name: When
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_005
    name: However
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_006
    name: Teachers
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_007
    name: Avoid
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_008
    name: Students
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
```

## 4. CONCEPT GRAPH

```yaml
concepts:
  - id: CON_001
    label: Students
    definition: Learning improves when students actively retrieve information instead of passively rereading.
    domain: education / learning science
    depends_on: []
    contradicts: []
    supports:
      - CON_002
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_002
    label: Rereading
    definition: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
    domain: education / learning science
    depends_on:
      - CON_001
    contradicts: []
    supports:
      - CON_003
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_003
    label: Memory
    definition: Sleep supports consolidation, while distraction reduces attention and weakens encoding.
    domain: education / learning science
    depends_on:
      - CON_002
    contradicts: []
    supports:
      - CON_004
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_004
    label: Sleep
    definition: When learners test themselves, they identify gaps that targeted review can close.
    domain: education / learning science
    depends_on:
      - CON_003
    contradicts: []
    supports:
      - CON_005
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_005
    label: Consolidation
    definition: If a concept is reviewed only once, it tends to fade quickly, because memory traces decay without reactivation.
    domain: education / learning science
    depends_on:
      - CON_004
    contradicts: []
    supports:
      - CON_006
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_006
    label: Because
    definition: However, if the spacing interval is too long, retrieval becomes effortful and accuracy drops.
    domain: education / learning science
    depends_on:
      - CON_005
    contradicts: []
    supports: []
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
```

## 5. PRINCIPLES

```yaml
principles: []  # none extracted
```

## 6. HEURISTICS

```yaml
heuristics:
  - id: HEU_001
    trigger: relevant decision context detected
    interpretation: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
  - id: HEU_002
    trigger: context contains a known failure mode
    interpretation: Avoid cramming the night before exams, because consolidation requires sleep.
    recommended_action: Avoid the described action.
    avoid: Avoid cramming the night before exams, because consolidation requires sleep.
    confidence: 0.76
```

## 7. DECISION RULES

```yaml
decision_rules:
  - id: RULE_001
    condition: Operating context matches the rule's domain.
    decision: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
```

## 8. PROCEDURES

```yaml
procedures:
  - id: PROC_001
    name: Source-derived procedure
    objective: Apply the sequence implied by the source text.
    steps:
      - step: 1
        action: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
        input_required: —
        output_expected: —
    success_criteria: All steps applied in order with expected outcomes.
    failure_criteria: Steps executed out of order or without prerequisites.
```

## 9. PATTERNS

```yaml
patterns:
  - id: PAT_001
    name: Recurring pattern 1
    observed_when: Source-described conditions are present.
    signal: If a concept is reviewed only once, it tends to fade quickly, because memory traces decay without reactivation.
    underlying_mechanism: —
    response_strategy: Recognize and act according to source guidance.
    confidence: 0.7
  - id: PAT_002
    name: Recurring pattern 2
    observed_when: Source-described conditions are present.
    signal: Students often confuse fluency with mastery, which leads to overconfidence.
    underlying_mechanism: —
    response_strategy: Recognize and act according to source guidance.
    confidence: 0.7
```

## 10. ANTI-PATTERNS

```yaml
anti_patterns:
  - id: ANTI_001
    name: Anti-pattern 1
    description: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes.
    why_it_fails: Identified by the source as ineffective or harmful.
    warning_signals: Behavior matches the described failure mode.
    replacement_behavior: Use the recommended alternative from the source.
  - id: ANTI_002
    name: Anti-pattern 2
    description: Avoid cramming the night before exams, because consolidation requires sleep.
    why_it_fails: Identified by the source as ineffective or harmful.
    warning_signals: Behavior matches the described failure mode.
    replacement_behavior: Use the recommended alternative from the source.
```

## 11. CAUSAL CHAINS

```yaml
causal_chains:
  - id: CAU_001
    cause: If a concept is reviewed only once, it tends to fade quickly,
    mechanism: —
    effect: memory traces decay without reactivation.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
  - id: CAU_002
    cause: Avoid cramming the night before exams,
    mechanism: —
    effect: consolidation requires sleep.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
  - id: CAU_003
    cause: Students often confuse fluency with mastery, which
    mechanism: —
    effect: overconfidence.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
```

## 12. CONTEXTUAL TRIGGERS

```yaml
contextual_triggers:
  - id: TRG_001
    if_user_says_or_context_contains: Students
    activate_knowledge:
      - CON_001
      - CON_002
    agent_should: Recall the Students concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_002
    if_user_says_or_context_contains: Rereading
    activate_knowledge:
      - CON_002
      - CON_003
    agent_should: Recall the Rereading concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_003
    if_user_says_or_context_contains: Memory
    activate_knowledge:
      - CON_003
      - CON_004
    agent_should: Recall the Memory concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
```

## 13. IF-THEN RULES

```yaml
if_then_rules:
  - id: IFTHEN_001
    if: students actively retrieve information instead of passively
    then: rereading
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_002
    if: learners test themselves
    then: they identify gaps that targeted review can close
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_003
    if: a concept is reviewed only once
    then: it tends to fade quickly, because memory traces decay without reactivation
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_004
    if: the spacing interval is too long
    then: retrieval becomes effortful and accuracy drops
    because: Inferred from source context.
    confidence: 0.72
```

## 14. EXCEPTIONS AND EDGE CASES

```yaml
exceptions:
  - id: EXC_001
    general_rule: —
    exception_case: However, if the spacing interval is too long, retrieval becomes effortful and accuracy drops.
    modified_action: Adjust behavior according to the exception.
    explanation: Source explicitly notes this edge case.
```

## 15. MENTAL MODELS

```yaml
mental_models:
  - id: MM_001
    name: Students model
    description: Learning improves when students actively retrieve information instead of passively rereading.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
  - id: MM_002
    name: Rereading model
    description: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
```

## 16. OPERATIONAL PLAYBOOKS

```yaml
playbooks:
  - id: PLAY_001
    name: education / learning science response playbook
    objective: Apply the source's knowledge to a real interaction.
    activation_context: User asks about Students.
    steps:
      - Identify the concept the question maps to.
      - Recall related rules and heuristics.
      - Cite the source-derived principle.
      - Surface relevant exceptions or limits.
    agent_tone: Clear, sourced, non-overstating.
    tools_needed:
      - retrieval
      - memory
    expected_output: A grounded answer with traceable reasoning.
    failure_modes:
      - Hallucinating beyond source
      - Ignoring exceptions
```

## 17. QUESTION-ANSWER PAIRS FOR AGENTS

```yaml
qa_pairs:
  - id: QA_001
    question: What is students and when does it apply?
    ideal_answer: Learning improves when students actively retrieve information instead of passively rereading.
    source_concepts:
      - CON_001
    difficulty: easy
    answer_type: definition_with_context
  - id: QA_002
    question: What is rereading and when does it apply?
    ideal_answer: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
    source_concepts:
      - CON_002
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_003
    question: What is memory and when does it apply?
    ideal_answer: Sleep supports consolidation, while distraction reduces attention and weakens encoding.
    source_concepts:
      - CON_003
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_004
    question: What is sleep and when does it apply?
    ideal_answer: When learners test themselves, they identify gaps that targeted review can close.
    source_concepts:
      - CON_004
    difficulty: medium
    answer_type: definition_with_context
```

## 18. RETRIEVAL CHUNKS

```yaml
retrieval_chunks:
  - id: CHUNK_001
    title: Chunk on learning
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Learning improves when students actively retrieve information instead of passively rereading. Spaced repetition strengthens memory by reactivating knowledge over increasing intervals. Sleep supports consolidation, while distraction reduces attention and weakens encoding.
    activation_queries:
      - What does the source say about learning?
      - What does the source say about improves?
      - What does the source say about students?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_002
    title: Chunk on learners
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: When learners test themselves, they identify gaps that targeted review can close. If a concept is reviewed only once, it tends to fade quickly, because memory traces decay without reactivation. However, if the spacing interval is too long, retrieval becomes effortful and accuracy drops.
    activation_queries:
      - What does the source say about learners?
      - What does the source say about test?
      - What does the source say about themselves?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_003
    title: Chunk on avoid
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes. Avoid cramming the night before exams, because consolidation requires sleep. Students often confuse fluency with mastery, which leads to overconfidence.
    activation_queries:
      - What does the source say about avoid?
      - What does the source say about teachers?
      - What does the source say about should?
    related_rules: []
    related_entities: []
    related_concepts: []
```

## 19. EMBEDDING-READY ATOMIC UNITS

```yaml
atomic_units:
  - id: AU_001
    statement: Learning improves when students actively retrieve information instead of passively rereading.
    type: fact
    tags:
      - learning
      - improves
      - students
    dependencies: []
    confidence: 0.78
  - id: AU_002
    statement: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
    type: fact
    tags:
      - spaced
      - repetition
      - strengthens
    dependencies: []
    confidence: 0.78
  - id: AU_003
    statement: Sleep supports consolidation, while distraction reduces attention and weakens encoding.
    type: fact
    tags:
      - sleep
      - supports
      - consolidation
    dependencies: []
    confidence: 0.78
  - id: AU_004
    statement: When learners test themselves, they identify gaps that targeted review can close.
    type: fact
    tags:
      - learners
      - test
      - themselves
    dependencies: []
    confidence: 0.78
  - id: AU_005
    statement: If a concept is reviewed only once, it tends to fade quickly, because memory traces decay without reactivation.
    type: definition
    tags:
      - concept
      - reviewed
      - only
    dependencies: []
    confidence: 0.78
  - id: AU_006
    statement: However, if the spacing interval is too long, retrieval becomes effortful and accuracy drops.
    type: definition
    tags:
      - however
      - spacing
      - interval
    dependencies: []
    confidence: 0.78
  - id: AU_007
    statement: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes.
    type: rule
    tags:
      - teachers
      - should
      - design
    dependencies: []
    confidence: 0.78
```

## 20. AGENT INSTRUCTIONS

```yaml
agent_instructions:
  behavior_rules:
    - Stay within the package's scope.
    - Cite the source-derived chunk or rule when answering.
    - Encourage active retrieval over passive review.
    - Suggest spaced repetition where appropriate.
  reasoning_rules:
    - Use causal chains and IF-THEN rules before improvising.
    - Combine concepts only when supports/depends_on relationships allow it.
  response_rules:
    - Be concise unless the user asks for depth.
    - Surface confidence and source basis.
  forbidden_behaviors:
    - Fabricating sources.
    - Restating the source as personal opinion.
    - Ignoring decision rules in favor of fluency.
  preferred_questions:
    - What does the source say about …?
    - Which rule applies to this situation?
    - What are the limits of this knowledge?
  tool_usage_guidance:
    - Use retrieval before generation.
    - Use memory to track conversational context.
```

## 21. KNOWLEDGE LIMITS

```yaml
knowledge_limits:
  missing_context:
    - Source date and authorship are not always provided.
  weakly_supported_claims: []
  assumptions_detected:
    - Heuristic compilation assumes the input text is self-contained.
  possible_biases:
    - Single-source perspective.
  outdated_sections: []
  needs_human_review:
    - Decision rules and exceptions before production use.
```

## 22. SOURCE TRACEABILITY

```yaml
source_traceability:
  - extracted_item_id: CON_001
    source_location: user_input
    source_excerpt: Learning improves when students actively retrieve information instead of passively rereading.
    extraction_type: explicit
  - extracted_item_id: CON_002
    source_location: user_input
    source_excerpt: Spaced repetition strengthens memory by reactivating knowledge over increasing intervals.
    extraction_type: explicit
  - extracted_item_id: CON_003
    source_location: user_input
    source_excerpt: Sleep supports consolidation, while distraction reduces attention and weakens encoding.
    extraction_type: explicit
  - extracted_item_id: CON_004
    source_location: user_input
    source_excerpt: When learners test themselves, they identify gaps that targeted review can close.
    extraction_type: explicit
  - extracted_item_id: CON_005
    source_location: user_input
    source_excerpt: If a concept is reviewed only once, it tends to fade quickly, because memory traces decay without reactivation.
    extraction_type: explicit
  - extracted_item_id: CON_006
    source_location: user_input
    source_excerpt: However, if the spacing interval is too long, retrieval becomes effortful and accuracy drops.
    extraction_type: explicit
  - extracted_item_id: HEU_001
    source_location: user_input
    source_excerpt: Teachers should design study sessions that interleave topics, avoid passive rereading, and include short, frequent quizzes.
    extraction_type: explicit
  - extracted_item_id: HEU_002
    source_location: user_input
    source_excerpt: Avoid cramming the night before exams, because consolidation requires sleep.
    extraction_type: explicit
  - extracted_item_id: IFTHEN_001
    source_location: user_input
    source_excerpt: IF students actively retrieve information instead of passively THEN rereading
    extraction_type: explicit
  - extracted_item_id: IFTHEN_002
    source_location: user_input
    source_excerpt: IF learners test themselves THEN they identify gaps that targeted review can close
    extraction_type: explicit
  - extracted_item_id: IFTHEN_003
    source_location: user_input
    source_excerpt: IF a concept is reviewed only once THEN it tends to fade quickly, because memory traces decay without reactivation
    extraction_type: explicit
  - extracted_item_id: IFTHEN_004
    source_location: user_input
    source_excerpt: IF the spacing interval is too long THEN retrieval becomes effortful and accuracy drops
    extraction_type: explicit
```

Business operations

ckf_demo_1778889423741

8 entities · 6 concepts · 0 principles
markdown
# CKF — KNOWLEDGE CONTEXT PACKAGE

package_id: ckf_demo_1778889423741
protocol_version: ckf-0.1
source_type: article
source_title: Untitled source
source_author: Unknown
domain: business
subdomains: [value, growth, customer, should]
language: en
created_at: 2026-05-15T23:57:03.741Z
compression_level: standard
human_readability: 0.7
ai_utility_score: 0.81

---

## 1. CORE INTENT

```yaml
core_intent:
  primary_purpose: Capture and structure the knowledge expressed in the source.
  intended_user: Developers, researchers and agents consuming structured knowledge.
  intended_agent_use: Retrieval, reasoning, tutoring, decision support.
  transformation_goal: Convert prose into structured, agent-usable cognition.
  key_value: Portable, traceable, reusable knowledge package.
```

## 2. DOMAIN MAP

```yaml
domain_map:
  main_domain: business
  subdomains:
    - name: value
      relevance: 1
      related_concepts:
        - value
    - name: growth
      relevance: 0.85
      related_concepts:
        - growth
    - name: customer
      relevance: 0.7
      related_concepts:
        - customer
    - name: should
      relevance: 0.55
      related_concepts:
        - should
  adjacent_domains: []
  excluded_domains: []
```

## 3. ENTITY GRAPH

```yaml
entities:
  - id: ENT_001
    name: Sustainable
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities: []
    source_basis: explicit
  - id: ENT_002
    name: Companies
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_003
    name: When
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_004
    name: Pricing
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_005
    name: Avoid
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_006
    name: Strong
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_007
    name: Leaders
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_008
    name: Sales
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
```

## 4. CONCEPT GRAPH

```yaml
concepts:
  - id: CON_001
    label: Value
    definition: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features.
    domain: business
    depends_on: []
    contradicts: []
    supports:
      - CON_002
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_002
    label: Growth
    definition: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    domain: business
    depends_on:
      - CON_001
    contradicts: []
    supports:
      - CON_003
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_003
    label: Customer
    definition: When customer acquisition cost exceeds lifetime value, expansion destroys value.
    domain: business
    depends_on:
      - CON_002
    contradicts: []
    supports:
      - CON_004
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_004
    label: Should
    definition: Pricing should reflect the value delivered, not internal costs alone.
    domain: business
    depends_on:
      - CON_003
    contradicts: []
    supports:
      - CON_005
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_005
    label: Before
    definition: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    domain: business
    depends_on:
      - CON_004
    contradicts: []
    supports:
      - CON_006
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_006
    label: Because
    definition: If churn is rising, revisit onboarding before optimizing acquisition.
    domain: business
    depends_on:
      - CON_005
    contradicts: []
    supports: []
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
```

## 5. PRINCIPLES

```yaml
principles: []  # none extracted
```

## 6. HEURISTICS

```yaml
heuristics:
  - id: HEU_001
    trigger: relevant decision context detected
    interpretation: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
  - id: HEU_002
    trigger: relevant decision context detected
    interpretation: Pricing should reflect the value delivered, not internal costs alone.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
  - id: HEU_003
    trigger: context contains a known failure mode
    interpretation: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    recommended_action: Avoid the described action.
    avoid: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    confidence: 0.76
  - id: HEU_004
    trigger: relevant decision context detected
    interpretation: Sales teams must qualify ruthlessly; closing poor-fit customers increases support load and drives downstream churn.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
```

## 7. DECISION RULES

```yaml
decision_rules:
  - id: RULE_001
    condition: Operating context matches the rule's domain.
    decision: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
  - id: RULE_002
    condition: Operating context matches the rule's domain.
    decision: Pricing should reflect the value delivered, not internal costs alone.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
  - id: RULE_003
    condition: Operating context matches the rule's domain.
    decision: Sales teams must qualify ruthlessly; closing poor-fit customers increases support load and drives downstream churn.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
```

## 8. PROCEDURES

```yaml
procedures: []  # none extracted
```

## 9. PATTERNS

```yaml
patterns:
  - id: PAT_001
    name: Recurring pattern 1
    observed_when: Source-described conditions are present.
    signal: Leaders often underinvest in retention, which leads to leaky funnels and stagnant revenue.
    underlying_mechanism: —
    response_strategy: Recognize and act according to source guidance.
    confidence: 0.7
```

## 10. ANTI-PATTERNS

```yaml
anti_patterns:
  - id: ANTI_001
    name: Anti-pattern 1
    description: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    why_it_fails: Identified by the source as ineffective or harmful.
    warning_signals: Behavior matches the described failure mode.
    replacement_behavior: Use the recommended alternative from the source.
```

## 11. CAUSAL CHAINS

```yaml
causal_chains:
  - id: CAU_001
    cause: Companies should measure unit economics before scaling marketing spend,
    mechanism: —
    effect: growth without margin amplifies losses.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
  - id: CAU_002
    cause: Avoid discounting as a default response to slow sales,
    mechanism: —
    effect: it trains the market to wait for promotions and erodes brand perception.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
  - id: CAU_003
    cause: Leaders often underinvest in retention, which
    mechanism: —
    effect: leaky funnels and stagnant revenue.
    secondary_effects: []
    intervention_points: []
    confidence: 0.7
```

## 12. CONTEXTUAL TRIGGERS

```yaml
contextual_triggers:
  - id: TRG_001
    if_user_says_or_context_contains: Value
    activate_knowledge:
      - CON_001
      - CON_002
    agent_should: Recall the Value concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_002
    if_user_says_or_context_contains: Growth
    activate_knowledge:
      - CON_002
      - CON_003
    agent_should: Recall the Growth concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_003
    if_user_says_or_context_contains: Customer
    activate_knowledge:
      - CON_003
      - CON_004
    agent_should: Recall the Customer concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
```

## 13. IF-THEN RULES

```yaml
if_then_rules:
  - id: IFTHEN_001
    if: customer acquisition cost exceeds lifetime value
    then: expansion destroys value
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_002
    if: churn is rising
    then: revisit onboarding before optimizing acquisition
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_003
    if: they consistently deliver on a clear
    then: promise
    because: Inferred from source context.
    confidence: 0.72
```

## 14. EXCEPTIONS AND EDGE CASES

```yaml
exceptions: []  # none extracted
```

## 15. MENTAL MODELS

```yaml
mental_models:
  - id: MM_001
    name: Value model
    description: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
  - id: MM_002
    name: Growth model
    description: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
```

## 16. OPERATIONAL PLAYBOOKS

```yaml
playbooks:
  - id: PLAY_001
    name: business response playbook
    objective: Apply the source's knowledge to a real interaction.
    activation_context: User asks about Value.
    steps:
      - Identify the concept the question maps to.
      - Recall related rules and heuristics.
      - Cite the source-derived principle.
      - Surface relevant exceptions or limits.
    agent_tone: Clear, sourced, non-overstating.
    tools_needed:
      - retrieval
      - memory
    expected_output: A grounded answer with traceable reasoning.
    failure_modes:
      - Hallucinating beyond source
      - Ignoring exceptions
```

## 17. QUESTION-ANSWER PAIRS FOR AGENTS

```yaml
qa_pairs:
  - id: QA_001
    question: What is value and when does it apply?
    ideal_answer: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features.
    source_concepts:
      - CON_001
    difficulty: easy
    answer_type: definition_with_context
  - id: QA_002
    question: What is growth and when does it apply?
    ideal_answer: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    source_concepts:
      - CON_002
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_003
    question: What is customer and when does it apply?
    ideal_answer: When customer acquisition cost exceeds lifetime value, expansion destroys value.
    source_concepts:
      - CON_003
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_004
    question: What is should and when does it apply?
    ideal_answer: Pricing should reflect the value delivered, not internal costs alone.
    source_concepts:
      - CON_004
    difficulty: medium
    answer_type: definition_with_context
```

## 18. RETRIEVAL CHUNKS

```yaml
retrieval_chunks:
  - id: CHUNK_001
    title: Chunk on growth
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features. Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses. When customer acquisition cost exceeds lifetime value, expansion destroys value.
    activation_queries:
      - What does the source say about growth?
      - What does the source say about customer?
      - What does the source say about value?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_002
    title: Chunk on pricing
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Pricing should reflect the value delivered, not internal costs alone. Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception. If churn is rising, revisit onboarding before optimizing acquisition.
    activation_queries:
      - What does the source say about pricing?
      - What does the source say about should?
      - What does the source say about reflect?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_003
    title: Chunk on strong
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Strong brands tend to compound over time when they consistently deliver on a clear promise. Leaders often underinvest in retention, which leads to leaky funnels and stagnant revenue. Sales teams must qualify ruthlessly; closing poor-fit customers increases support load and drives downstream churn.
    activation_queries:
      - What does the source say about strong?
      - What does the source say about brands?
      - What does the source say about tend?
    related_rules: []
    related_entities: []
    related_concepts: []
```

## 19. EMBEDDING-READY ATOMIC UNITS

```yaml
atomic_units:
  - id: AU_001
    statement: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features.
    type: fact
    tags:
      - sustainable
      - business
      - growth
    dependencies: []
    confidence: 0.78
  - id: AU_002
    statement: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    type: rule
    tags:
      - companies
      - should
      - measure
    dependencies: []
    confidence: 0.78
  - id: AU_003
    statement: When customer acquisition cost exceeds lifetime value, expansion destroys value.
    type: fact
    tags:
      - value
      - customer
      - acquisition
    dependencies: []
    confidence: 0.78
  - id: AU_004
    statement: Pricing should reflect the value delivered, not internal costs alone.
    type: rule
    tags:
      - pricing
      - should
      - reflect
    dependencies: []
    confidence: 0.78
  - id: AU_005
    statement: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    type: rule
    tags:
      - avoid
      - discounting
      - default
    dependencies: []
    confidence: 0.78
  - id: AU_006
    statement: If churn is rising, revisit onboarding before optimizing acquisition.
    type: definition
    tags:
      - churn
      - rising
      - revisit
    dependencies: []
    confidence: 0.78
  - id: AU_007
    statement: Strong brands tend to compound over time when they consistently deliver on a clear promise.
    type: fact
    tags:
      - strong
      - brands
      - tend
    dependencies: []
    confidence: 0.78
```

## 20. AGENT INSTRUCTIONS

```yaml
agent_instructions:
  behavior_rules:
    - Stay within the package's scope.
    - Cite the source-derived chunk or rule when answering.
    - Distinguish between strategy and tactics.
    - Surface assumptions about the market.
  reasoning_rules:
    - Use causal chains and IF-THEN rules before improvising.
    - Combine concepts only when supports/depends_on relationships allow it.
  response_rules:
    - Be concise unless the user asks for depth.
    - Surface confidence and source basis.
  forbidden_behaviors:
    - Fabricating sources.
    - Restating the source as personal opinion.
    - Ignoring decision rules in favor of fluency.
  preferred_questions:
    - What does the source say about …?
    - Which rule applies to this situation?
    - What are the limits of this knowledge?
  tool_usage_guidance:
    - Use retrieval before generation.
    - Use memory to track conversational context.
```

## 21. KNOWLEDGE LIMITS

```yaml
knowledge_limits:
  missing_context:
    - Source date and authorship are not always provided.
  weakly_supported_claims: []
  assumptions_detected:
    - Heuristic compilation assumes the input text is self-contained.
  possible_biases:
    - Single-source perspective.
  outdated_sections: []
  needs_human_review:
    - Decision rules and exceptions before production use.
```

## 22. SOURCE TRACEABILITY

```yaml
source_traceability:
  - extracted_item_id: CON_001
    source_location: user_input
    source_excerpt: Sustainable business growth depends on understanding the customer's underlying job to be done, not only the product features.
    extraction_type: explicit
  - extracted_item_id: CON_002
    source_location: user_input
    source_excerpt: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    extraction_type: explicit
  - extracted_item_id: CON_003
    source_location: user_input
    source_excerpt: When customer acquisition cost exceeds lifetime value, expansion destroys value.
    extraction_type: explicit
  - extracted_item_id: CON_004
    source_location: user_input
    source_excerpt: Pricing should reflect the value delivered, not internal costs alone.
    extraction_type: explicit
  - extracted_item_id: CON_005
    source_location: user_input
    source_excerpt: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    extraction_type: explicit
  - extracted_item_id: CON_006
    source_location: user_input
    source_excerpt: If churn is rising, revisit onboarding before optimizing acquisition.
    extraction_type: explicit
  - extracted_item_id: HEU_001
    source_location: user_input
    source_excerpt: Companies should measure unit economics before scaling marketing spend, because growth without margin amplifies losses.
    extraction_type: explicit
  - extracted_item_id: HEU_002
    source_location: user_input
    source_excerpt: Pricing should reflect the value delivered, not internal costs alone.
    extraction_type: explicit
  - extracted_item_id: HEU_003
    source_location: user_input
    source_excerpt: Avoid discounting as a default response to slow sales, because it trains the market to wait for promotions and erodes brand perception.
    extraction_type: explicit
  - extracted_item_id: HEU_004
    source_location: user_input
    source_excerpt: Sales teams must qualify ruthlessly; closing poor-fit customers increases support load and drives downstream churn.
    extraction_type: explicit
  - extracted_item_id: IFTHEN_001
    source_location: user_input
    source_excerpt: IF customer acquisition cost exceeds lifetime value THEN expansion destroys value
    extraction_type: explicit
  - extracted_item_id: IFTHEN_002
    source_location: user_input
    source_excerpt: IF churn is rising THEN revisit onboarding before optimizing acquisition
    extraction_type: explicit
  - extracted_item_id: IFTHEN_003
    source_location: user_input
    source_excerpt: IF they consistently deliver on a clear THEN promise
    extraction_type: explicit
```

Software engineering

ckf_demo_1778889423741

8 entities · 6 concepts · 0 principles
markdown
# CKF — KNOWLEDGE CONTEXT PACKAGE

package_id: ckf_demo_1778889423741
protocol_version: ckf-0.1
source_type: policy
source_title: Untitled source
source_author: Unknown
domain: business
subdomains: [should, security, before, change]
language: en
created_at: 2026-05-15T23:57:03.741Z
compression_level: standard
human_readability: 0.7
ai_utility_score: 0.81

---

## 1. CORE INTENT

```yaml
core_intent:
  primary_purpose: Capture and structure the knowledge expressed in the source.
  intended_user: Developers, researchers and agents consuming structured knowledge.
  intended_agent_use: Retrieval, reasoning, tutoring, decision support.
  transformation_goal: Convert prose into structured, agent-usable cognition.
  key_value: Portable, traceable, reusable knowledge package.
```

## 2. DOMAIN MAP

```yaml
domain_map:
  main_domain: business
  subdomains:
    - name: should
      relevance: 1
      related_concepts:
        - should
    - name: security
      relevance: 0.85
      related_concepts:
        - security
    - name: before
      relevance: 0.7
      related_concepts:
        - before
    - name: change
      relevance: 0.55
      related_concepts:
        - change
  adjacent_domains: []
  excluded_domains: []
```

## 3. ENTITY GRAPH

```yaml
entities:
  - id: ENT_001
    name: Production
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities: []
    source_basis: explicit
  - id: ENT_002
    name: Reviewers
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_003
    name: When
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_004
    name: Avoid
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_005
    name: Small
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_006
    name: Authors
    type: named_entity
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_007
    name: Should
    type: concept_term
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
  - id: ENT_008
    name: Security
    type: concept_term
    description: Recurring term/entity surfaced from the source.
    aliases: []
    attributes: []
    related_entities:
      - entity_id: ENT_001
        relation_type: co_occurs_with
        confidence: 0.6
    source_basis: explicit
```

## 4. CONCEPT GRAPH

```yaml
concepts:
  - id: CON_001
    label: Should
    definition: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    domain: business
    depends_on: []
    contradicts: []
    supports:
      - CON_002
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_002
    label: Security
    definition: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    domain: business
    depends_on:
      - CON_001
    contradicts: []
    supports:
      - CON_003
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_003
    label: Before
    definition: When tests are missing for a non-trivial change, request them before approving.
    domain: business
    depends_on:
      - CON_002
    contradicts: []
    supports:
      - CON_004
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_004
    label: Change
    definition: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    domain: business
    depends_on:
      - CON_003
    contradicts: []
    supports:
      - CON_005
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_005
    label: Pull
    definition: If a change touches a security boundary, require a second reviewer with security context.
    domain: business
    depends_on:
      - CON_004
    contradicts: []
    supports:
      - CON_006
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
  - id: CON_006
    label: Requests
    definition: Small, frequent pull requests get reviewed faster and merged with fewer regressions than large multi-feature branches.
    domain: business
    depends_on:
      - CON_005
    contradicts: []
    supports: []
    enables: []
    risks: []
    confidence: 0.78
    source_basis: explicit
```

## 5. PRINCIPLES

```yaml
principles: []  # none extracted
```

## 6. HEURISTICS

```yaml
heuristics:
  - id: HEU_001
    trigger: relevant decision context detected
    interpretation: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
  - id: HEU_002
    trigger: relevant decision context detected
    interpretation: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
  - id: HEU_003
    trigger: context contains a known failure mode
    interpretation: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    recommended_action: Avoid the described action.
    avoid: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    confidence: 0.76
  - id: HEU_004
    trigger: relevant decision context detected
    interpretation: Authors should keep the description focused on intent and risk, not on a step-by-step diff narration.
    recommended_action: Follow the recommended practice.
    avoid: —
    confidence: 0.76
```

## 7. DECISION RULES

```yaml
decision_rules:
  - id: RULE_001
    condition: Operating context matches the rule's domain.
    decision: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
  - id: RULE_002
    condition: Operating context matches the rule's domain.
    decision: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
  - id: RULE_003
    condition: Operating context matches the rule's domain.
    decision: Authors should keep the description focused on intent and risk, not on a step-by-step diff narration.
    reasoning: Derived directly from a normative statement in the source.
    required_context: Domain-specific context as described in the source.
    output_action: Apply the recommended decision.
    failure_mode: Recommendation applied outside its valid context.
    confidence: 0.74
```

## 8. PROCEDURES

```yaml
procedures: []  # none extracted
```

## 9. PATTERNS

```yaml
patterns: []  # none extracted
```

## 10. ANTI-PATTERNS

```yaml
anti_patterns:
  - id: ANTI_001
    name: Anti-pattern 1
    description: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    why_it_fails: Identified by the source as ineffective or harmful.
    warning_signals: Behavior matches the described failure mode.
    replacement_behavior: Use the recommended alternative from the source.
```

## 11. CAUSAL CHAINS

```yaml
causal_chains: []  # none extracted
```

## 12. CONTEXTUAL TRIGGERS

```yaml
contextual_triggers:
  - id: TRG_001
    if_user_says_or_context_contains: Should
    activate_knowledge:
      - CON_001
      - CON_002
    agent_should: Recall the Should concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_002
    if_user_says_or_context_contains: Security
    activate_knowledge:
      - CON_002
      - CON_003
    agent_should: Recall the Security concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
  - id: TRG_003
    if_user_says_or_context_contains: Before
    activate_knowledge:
      - CON_003
      - CON_004
    agent_should: Recall the Before concept and apply related rules.
    agent_should_not: Make claims beyond what the source supports.
```

## 13. IF-THEN RULES

```yaml
if_then_rules:
  - id: IFTHEN_001
    if: tests are missing for a non-trivial change
    then: request them before approving
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_002
    if: a linter could enforce the rule
    then: automatically
    because: Inferred from source context.
    confidence: 0.72
  - id: IFTHEN_003
    if: a change touches a security boundary
    then: require a second reviewer with security context
    because: Inferred from source context.
    confidence: 0.72
```

## 14. EXCEPTIONS AND EDGE CASES

```yaml
exceptions: []  # none extracted
```

## 15. MENTAL MODELS

```yaml
mental_models:
  - id: MM_001
    name: Should model
    description: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
  - id: MM_002
    name: Security model
    description: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    use_when: The reasoning context maps to this concept.
    do_not_use_when: Context lies outside the source's scope.
    input_needed: Relevant facts about the situation.
    output_generated: A reasoned recommendation aligned with the source.
```

## 16. OPERATIONAL PLAYBOOKS

```yaml
playbooks:
  - id: PLAY_001
    name: business response playbook
    objective: Apply the source's knowledge to a real interaction.
    activation_context: User asks about Should.
    steps:
      - Identify the concept the question maps to.
      - Recall related rules and heuristics.
      - Cite the source-derived principle.
      - Surface relevant exceptions or limits.
    agent_tone: Clear, sourced, non-overstating.
    tools_needed:
      - retrieval
      - memory
    expected_output: A grounded answer with traceable reasoning.
    failure_modes:
      - Hallucinating beyond source
      - Ignoring exceptions
```

## 17. QUESTION-ANSWER PAIRS FOR AGENTS

```yaml
qa_pairs:
  - id: QA_001
    question: What is should and when does it apply?
    ideal_answer: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    source_concepts:
      - CON_001
    difficulty: easy
    answer_type: definition_with_context
  - id: QA_002
    question: What is security and when does it apply?
    ideal_answer: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    source_concepts:
      - CON_002
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_003
    question: What is before and when does it apply?
    ideal_answer: When tests are missing for a non-trivial change, request them before approving.
    source_concepts:
      - CON_003
    difficulty: medium
    answer_type: definition_with_context
  - id: QA_004
    question: What is change and when does it apply?
    ideal_answer: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    source_concepts:
      - CON_004
    difficulty: medium
    answer_type: definition_with_context
```

## 18. RETRIEVAL CHUNKS

```yaml
retrieval_chunks:
  - id: CHUNK_001
    title: Chunk on should
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Production code reviews should optimize for catching defects early, not for enforcing personal style. Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    activation_queries:
      - What does the source say about should?
      - What does the source say about production?
      - What does the source say about code?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_002
    title: Chunk on tests
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: When tests are missing for a non-trivial change, request them before approving. Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    activation_queries:
      - What does the source say about tests?
      - What does the source say about missing?
      - What does the source say about non-trivial?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_003
    title: Chunk on security
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: If a change touches a security boundary, require a second reviewer with security context. Small, frequent pull requests get reviewed faster and merged with fewer regressions than large multi-feature branches.
    activation_queries:
      - What does the source say about security?
      - What does the source say about change?
      - What does the source say about touches?
    related_rules: []
    related_entities: []
    related_concepts: []
  - id: CHUNK_004
    title: Chunk on authors
    standalone_context: Self-contained passage extracted from the source.
    compressed_knowledge: Authors should keep the description focused on intent and risk, not on a step-by-step diff narration.
    activation_queries:
      - What does the source say about authors?
      - What does the source say about should?
      - What does the source say about keep?
    related_rules: []
    related_entities: []
    related_concepts: []
```

## 19. EMBEDDING-READY ATOMIC UNITS

```yaml
atomic_units:
  - id: AU_001
    statement: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    type: rule
    tags:
      - production
      - code
      - reviews
    dependencies: []
    confidence: 0.78
  - id: AU_002
    statement: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    type: rule
    tags:
      - reviewers
      - should
      - focus
    dependencies: []
    confidence: 0.78
  - id: AU_003
    statement: When tests are missing for a non-trivial change, request them before approving.
    type: definition
    tags:
      - tests
      - missing
      - non-trivial
    dependencies: []
    confidence: 0.78
  - id: AU_004
    statement: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    type: rule
    tags:
      - avoid
      - blocking
      - pull
    dependencies: []
    confidence: 0.78
  - id: AU_005
    statement: If a change touches a security boundary, require a second reviewer with security context.
    type: fact
    tags:
      - security
      - change
      - touches
    dependencies: []
    confidence: 0.78
  - id: AU_006
    statement: Small, frequent pull requests get reviewed faster and merged with fewer regressions than large multi-feature branches.
    type: fact
    tags:
      - small
      - frequent
      - pull
    dependencies: []
    confidence: 0.78
  - id: AU_007
    statement: Authors should keep the description focused on intent and risk, not on a step-by-step diff narration.
    type: rule
    tags:
      - authors
      - should
      - keep
    dependencies: []
    confidence: 0.78
```

## 20. AGENT INSTRUCTIONS

```yaml
agent_instructions:
  behavior_rules:
    - Stay within the package's scope.
    - Cite the source-derived chunk or rule when answering.
    - Distinguish between strategy and tactics.
    - Surface assumptions about the market.
  reasoning_rules:
    - Use causal chains and IF-THEN rules before improvising.
    - Combine concepts only when supports/depends_on relationships allow it.
  response_rules:
    - Be concise unless the user asks for depth.
    - Surface confidence and source basis.
  forbidden_behaviors:
    - Fabricating sources.
    - Restating the source as personal opinion.
    - Ignoring decision rules in favor of fluency.
  preferred_questions:
    - What does the source say about …?
    - Which rule applies to this situation?
    - What are the limits of this knowledge?
  tool_usage_guidance:
    - Use retrieval before generation.
    - Use memory to track conversational context.
```

## 21. KNOWLEDGE LIMITS

```yaml
knowledge_limits:
  missing_context:
    - Source date and authorship are not always provided.
  weakly_supported_claims: []
  assumptions_detected:
    - Heuristic compilation assumes the input text is self-contained.
  possible_biases:
    - Single-source perspective.
  outdated_sections: []
  needs_human_review:
    - Decision rules and exceptions before production use.
```

## 22. SOURCE TRACEABILITY

```yaml
source_traceability:
  - extracted_item_id: CON_001
    source_location: user_input
    source_excerpt: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    extraction_type: explicit
  - extracted_item_id: CON_002
    source_location: user_input
    source_excerpt: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    extraction_type: explicit
  - extracted_item_id: CON_003
    source_location: user_input
    source_excerpt: When tests are missing for a non-trivial change, request them before approving.
    extraction_type: explicit
  - extracted_item_id: CON_004
    source_location: user_input
    source_excerpt: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    extraction_type: explicit
  - extracted_item_id: CON_005
    source_location: user_input
    source_excerpt: If a change touches a security boundary, require a second reviewer with security context.
    extraction_type: explicit
  - extracted_item_id: CON_006
    source_location: user_input
    source_excerpt: Small, frequent pull requests get reviewed faster and merged with fewer regressions than large multi-feature branches.
    extraction_type: explicit
  - extracted_item_id: HEU_001
    source_location: user_input
    source_excerpt: Production code reviews should optimize for catching defects early, not for enforcing personal style.
    extraction_type: explicit
  - extracted_item_id: HEU_002
    source_location: user_input
    source_excerpt: Reviewers should focus on correctness, security, and architectural fit before line-by-line nitpicks.
    extraction_type: explicit
  - extracted_item_id: HEU_003
    source_location: user_input
    source_excerpt: Avoid blocking pull requests on subjective formatting when a linter could enforce the rule automatically.
    extraction_type: explicit
  - extracted_item_id: HEU_004
    source_location: user_input
    source_excerpt: Authors should keep the description focused on intent and risk, not on a step-by-step diff narration.
    extraction_type: explicit
  - extracted_item_id: IFTHEN_001
    source_location: user_input
    source_excerpt: IF tests are missing for a non-trivial change THEN request them before approving
    extraction_type: explicit
  - extracted_item_id: IFTHEN_002
    source_location: user_input
    source_excerpt: IF a linter could enforce the rule THEN automatically
    extraction_type: explicit
  - extracted_item_id: IFTHEN_003
    source_location: user_input
    source_excerpt: IF a change touches a security boundary THEN require a second reviewer with security context
    extraction_type: explicit
```