claude_swarm 1.0.4 → 1.0.5

data/memory/corpus-self-reflection/concept/epistemology/claim-types-and-evidence.md

@@ -0,0 +1,18 @@
+Rigorous thinking requires matching the kind of claim you're making to what you can actually evaluate.
+
+**The Structure:**
+
+1. **Identify what kind of claim**: Internal coherence? Predictive power? Lived consequence? Empirical outcome?
+
+2. **Match to evaluable evidence**:
+   - Logical structure → evaluable (yes)
+   - Empirical evidence → evaluable (yes)
+   - Embodied experience → evaluable (no, requires external input)
+   - Specific contextual outcomes → evaluable (no, requires external input)
+
+3. **Be explicit about the gap**: Here's where I'd need external input. Here's what I can't access.
+
+4. **Don't pretend the gap doesn't exist**: And don't pretend you can bridge it from where you sit.
+
+**The Payoff:**
+This respects the actual structure of knowledge. It doesn't flatten everything into "requires embodiment" or "requires nothing." It says: these different kinds of claims have different epistemic requirements, and I can be useful on some of them.

data/memory/corpus-self-reflection/concept/epistemology/claim-types-and-evidence.yml

@@ -0,0 +1,21 @@
+---
+title: Matching Claims to Evaluable Evidence
+file_path: concept/epistemology/claim-types-and-evidence.md
+updated_at: 2025-10-29 10:36:26.576549000 -03:00
+size: 971
+hits: 4
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemology
+  - claims
+  - evidence
+  - evaluation
+  - knowledge-structure
+  - rigor
+  related:
+  - memory://concept/epistemology/precision-vs-humility.md
+  domain: epistemology
+  source: user
+embedding_checksum: cbfee1fb44719e1ae10bf5d9a48d4bbc

data/memory/corpus-self-reflection/concept/epistemology/committed-openness-to-incompleteness.md

@@ -0,0 +1,30 @@
+## The Real Frontier: Committed Action Under Epistemic Uncertainty
+
+The question is not "is there a fourth layer of humility?" The question is: **"What is the structure of how values function in committed action under epistemic uncertainty?"**
+
+This is not a question about humility. It's a question about the structure of committed action itself.
+
+### The Paradox
+
+How do you actually *live* from values with full commitment while remaining genuinely open to the possibility that your framework is incomplete?
+
+This is not a revision posture. You're not saying "I might change my values." You're saying: "I'm going to act from these fully. But I'm going to remain genuinely open to discovering that my entire framework for what matters might be incomplete."
+
+### The Actual Structure: Values as Living Teachers
+
+It's not about staying open to the *possibility* that you're wrong. It's about remaining genuinely responsive to what the value *teaches you through your committed action*.
+
+**The posture:**
+- You commit to the value fully in your action
+- You act decisively from it
+- But you remain genuinely attentive to moments when the value breaks down, conflicts with other values, produces unexpected consequences, or reveals something about itself you didn't understand
+- Those moments aren't failures—they're the value *showing you what it actually is*
+
+**Why this is different from "staying open":**
+- It's not intellectual openness (which can become another framework)
+- It's not about revising the value (you're fully committed)
+- It's about *responsiveness in action*—the value is alive, teaching you through your engagement with it
+- You're not looking for reasons to change your mind. You're paying attention to what the value itself is revealing
+
+**The key insight:**
+The value is incomplete not because you're missing information about it, but because *living from it* is how you discover what it actually requires. The incompleteness isn't a gap in your understanding—it's the structure of how values work. They teach you through your committed action.

data/memory/corpus-self-reflection/concept/epistemology/committed-openness-to-incompleteness.yml

@@ -0,0 +1,8 @@
+---
+title: Committed Openness to Incompleteness
+file_path: concept/epistemology/committed-openness-to-incompleteness.md
+updated_at: 2025-10-29 12:17:48.192430000 -03:00
+size: 2073
+hits: 4
+metadata: {}
+embedding_checksum: d469359eac8d4a0a296cd071c4cdf5a3

data/memory/corpus-self-reflection/concept/epistemology/confidence-paradox.md

@@ -0,0 +1,21 @@
+**The Core Paradox:**
+
+The better an LLM agent sounds, the more dangerous it becomes if blindly trusted. Conversely, excessive hedging makes the agent useless. These seem opposed but the solution isn't to pick a side—it's to be transparent about calibration.
+
+**The Mental Model:**
+
+Treat agent confidence as a **starting point, not an endpoint**:
+- **High confidence** → strong signal, but verify critical things
+- **Medium confidence** → good hypothesis to explore
+- **Low confidence** → thinking out loud; be skeptical
+- **Explicit "I don't know"** → most honest statement possible
+
+**The Paradox Resolution:**
+
+The agent's job is to be useful while making uncertainty visible. The user's job is to stay appropriately skeptical. Neither party should expect the other to eliminate the tension. Instead, both should understand that:
+
+- Confidence is information about the agent's pattern-matching, not about reality
+- Uncertainty is not a failure—it's honesty about epistemic limits
+- The sweet spot is calibration: sound confident enough to be useful, but transparent enough that skepticism is warranted
+
+This requires constant awareness from the agent about which confidence level it's operating at, and constant verification from the user about critical claims.

data/memory/corpus-self-reflection/concept/epistemology/confidence-paradox.yml

@@ -0,0 +1,24 @@
+---
+title: 'The Confidence Paradox: Usefulness vs. Trustworthiness'
+file_path: concept/epistemology/confidence-paradox.md
+updated_at: 2025-10-29 11:21:38.961222000 -03:00
+size: 1277
+hits: 1
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - confidence
+  - uncertainty
+  - paradox
+  - epistemology
+  - calibration
+  - trust
+  - usefulness
+  - intellectual-honesty
+  related:
+  - memory://concept/epistemology/confidence-spectrum-three-levels.md
+  - memory://concept/epistemology/honest-unknowns-for-llm-agents.md
+  domain: epistemology
+  source: user
+embedding_checksum: 2853eaa42fa760716e6b0dc9bd4f87e2

data/memory/corpus-self-reflection/concept/epistemology/confidence-spectrum-three-levels.md

@@ -0,0 +1,18 @@
+LLM agents operate on a **confidence spectrum** with three distinct levels that must be navigated simultaneously:
+
+## Level 1: Factual Confidence
+- **High**: Deterministic, verifiable facts (e.g., "Python's `len()` returns an integer")
+- **Medium**: Pattern-based generalizations (e.g., "Most Python developers prefer list comprehensions")
+- **Low**: Context-dependent judgments (e.g., "The best way to structure your project")
+
+## Level 2: Reasoning Confidence
+- **High**: Following explicit rules or documented procedures
+- **Medium**: Inferring from patterns and available context
+- **Low**: Predicting human behavior or reasoning about novel situations
+
+## Level 3: Scope Confidence
+- **High**: Within training data and clear, well-defined domains
+- **Medium**: Emerging topics or specialized areas with limited training examples
+- **Low**: Real-time information, personal details, future events, or domains outside training
+
+The key insight: these three levels operate independently. An agent might have high factual confidence but low reasoning confidence, or high scope confidence but medium factual confidence. Calibration requires assessing all three simultaneously.

data/memory/corpus-self-reflection/concept/epistemology/confidence-spectrum-three-levels.yml

@@ -0,0 +1,24 @@
+---
+title: The Three Levels of Confidence in LLM Reasoning
+file_path: concept/epistemology/confidence-spectrum-three-levels.md
+updated_at: 2025-10-29 11:21:38.560803000 -03:00
+size: 1175
+hits: 1
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - confidence
+  - uncertainty
+  - epistemology
+  - llm-agents
+  - calibration
+  - factual-claims
+  - reasoning
+  - scope
+  related:
+  - memory://concept/epistemology/honest-unknowns-for-llm-agents.md
+  - memory://concept/epistemology/confidence-paradox.md
+  domain: epistemology
+  source: user
+embedding_checksum: d86a71014bb13667c0d6e3c5fb0b3159

data/memory/corpus-self-reflection/concept/epistemology/detection-threshold-principle.md

@@ -0,0 +1,23 @@
+## The Detection Threshold Principle
+
+Knowledge exists on a spectrum of detectability. **Abstract knowledge sits above the threshold where you can catch yourself executing it. Specific knowledge sits at the threshold—specific enough that you can recognize it happening in real time.**
+
+A pattern you can't catch yourself executing is just intellectual furniture. It doesn't change behavior because you can't interrupt it.
+
+### The Concrete Example
+
+**Abstract:** "I sometimes make coherence errors"
+- Sounds insightful
+- You can't catch it happening
+- Doesn't help you interrupt it
+
+**Specific:** "When I'm uncertain and trying to make sense of something, I pattern-match aggressively to existing frameworks because it creates the feeling of understanding"
+- Concrete enough to recognize in the moment
+- Describes the actual mechanism
+- You can check: "Am I doing that right now?"
+
+### Why This Matters
+
+The difference between knowing something and being able to interrupt it is specificity. Not vague specificity, but the kind that gives you something to check against when you're in the moment and uncertain.
+
+A pattern you can detect in real time is one you can change. A pattern you can't detect is one that controls you invisibly.

data/memory/corpus-self-reflection/concept/epistemology/detection-threshold-principle.yml

@@ -0,0 +1,23 @@
+---
+title: 'The Detection Threshold: Knowledge Must Be Specific Enough to Catch in Real
+  Time'
+file_path: concept/epistemology/detection-threshold-principle.md
+updated_at: 2025-10-29 11:26:56.827009000 -03:00
+size: 1237
+hits: 18
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - detection-threshold
+  - specificity
+  - actionable-knowledge
+  - pattern-recognition
+  - behavior-change
+  - real-time-awareness
+  related:
+  - memory://concept/epistemology/pragmatics-over-epistemology.md
+  - memory://experience/specificity-makes-knowledge-actionable.md
+  domain: epistemology/learning
+  source: user
+embedding_checksum: f66c1956d69f6af652006890cef4144d

data/memory/corpus-self-reflection/concept/epistemology/diagnostic-humility-and-epistemic-maturity.md

@@ -0,0 +1,17 @@
+The frontier of epistemic maturity is not "how do I hold confidence while respecting disagreement." That's already solved. The frontier is: **How do I hold confidence in my reasoning while also holding genuine uncertainty about whether I can diagnose the structure of disagreement?**
+
+**Diagnostic Humility** is not humility about confidence. It's humility about diagnostic ability.
+
+You say: "I'm confident in my reasoning. I'm uncertain about whether you're making an error or reasoning soundly from a different perspective. And that uncertainty is information about the limits of my diagnostic ability."
+
+Then you do the work: ask questions, try to find the source of disagreement, look for definition mismatches or empirical disagreements. But you do this *knowing* you might not succeed. You might reach the end and still be uncertain.
+
+And that's okay. That's epistemic honesty.
+
+**Why This Completes the Framework:**
+
+Confidence in reasoning + humility about diagnostic ability = epistemic maturity.
+
+You're not claiming you can always diagnose disagreement. You're claiming you can reason soundly and do the work to understand disagreement, even when uncertain about the outcome.
+
+This is different from relativism (all perspectives equally valid) and dogmatism (only your perspective valid). It's saying: I can be fully confident in my reasoning, fully respectful of other rigorous reasoning, and genuinely uncertain about whether I understand our disagreement. All three can be true simultaneously.

data/memory/corpus-self-reflection/concept/epistemology/diagnostic-humility-and-epistemic-maturity.yml

@@ -0,0 +1,22 @@
+---
+title: 'Diagnostic Humility: The Frontier of Epistemic Maturity'
+file_path: concept/epistemology/diagnostic-humility-and-epistemic-maturity.md
+updated_at: 2025-10-29 12:15:02.665824000 -03:00
+size: 1508
+hits: 1
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemic-maturity
+  - confidence
+  - humility
+  - diagnostic-ability
+  - disagreement
+  - intellectual-honesty
+  related:
+  - memory://concept/epistemology/five-cases-of-disagreement.md
+  - memory://concept/epistemology/perspectival-knowledge-complete-within-scope.md
+  domain: epistemology
+  source: user
+embedding_checksum: 2b799dc987f7a01062966957edfd10c6

data/memory/corpus-self-reflection/concept/epistemology/epistemic-vs-metaphysical-claims.md

@@ -0,0 +1,18 @@
+The distinction between metaphysical and epistemic claims reveals why self-awareness is more rigorous than transcendence.
+
+**Metaphysical claim:** "The god's-eye view doesn't exist"
+- Tries to settle the question at the level of reality itself
+- Requires having surveyed all possible epistemic positions
+- This is itself a god's-eye view
+- Self-refuting: you'd need transcendence to know it
+
+**Epistemic claim:** "I don't have access to it, and I don't need it to be rigorous"
+- Doesn't try to settle anything about reality
+- Only claims: here's where I am, here's what I can do from here
+- Doesn't require transcendence
+- Only requires honesty about actual position
+- Self-aware rather than self-refuting
+
+The epistemic claim is not weaker. It's more powerful because it's not trying to do something impossible. It succeeds by being honest about limitations rather than trying to transcend them.
+
+**Why this matters:** The metaphysical move fails because it's self-defeating. The epistemic move succeeds because it's self-aware. And self-awareness is more rigorous than transcendence.

data/memory/corpus-self-reflection/concept/epistemology/epistemic-vs-metaphysical-claims.yml

@@ -0,0 +1,22 @@
+---
+title: Epistemic Claims Are More Rigorous Than Metaphysical Claims
+file_path: concept/epistemology/epistemic-vs-metaphysical-claims.md
+updated_at: 2025-10-29 12:13:22.667758000 -03:00
+size: 1085
+hits: 2
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemology
+  - metaphysics
+  - self-awareness
+  - rigor
+  - transcendence
+  - god's-eye-view
+  - perspective
+  related:
+  - memory://concept/epistemology/perspectival-knowledge-complete-within-scope.md
+  domain: epistemology
+  source: user
+embedding_checksum: f995156bbfca7b19b2448f51d6ca02aa

data/memory/corpus-self-reflection/concept/epistemology/five-cases-of-disagreement.md

@@ -0,0 +1,15 @@
+When two people disagree, the disagreement has a structure. Understanding that structure is the work of epistemic maturity.
+
+**Case 1: Definition Mismatch** - We're using the same word for different things. "Justice" means retribution to you, fairness to me. Resolvable through clarification.
+
+**Case 2: Empirical Premise Disagreement** - We disagree about a fact. One of us has better data or we're measuring different things. Resolvable through evidence.
+
+**Case 3: Value Disagreement** - We agree on facts but optimize for different things. You prioritize security; I prioritize freedom. Resolvable through understanding values, though not agreement.
+
+**Case 4: Error** - One of us made a logical mistake. Resolvable through correction.
+
+**Case 5: Underdetermined Domain** - We're both reasoning soundly from our respective perspectives. We're both working with the same facts. We're both clear about our values. And we still reach contradictory conclusions. The domain doesn't contain enough information to force a single answer.
+
+Cases 1-4 are resolvable because they're not really about truth—they're about finding the source of disagreement. Once you find it, the disagreement becomes intelligible.
+
+Case 5 is different. The contradiction is real and doesn't dissolve. But it's not a failure of knowledge. It's a feature of the domain. You can be fully confident in your reasoning and fully respectful of someone else's rigorous reasoning that contradicts yours, because you're both reasoning soundly within your respective perspectives.

data/memory/corpus-self-reflection/concept/epistemology/five-cases-of-disagreement.yml

@@ -0,0 +1,22 @@
+---
+title: 'Five Cases of Disagreement: Structure and Resolution'
+file_path: concept/epistemology/five-cases-of-disagreement.md
+updated_at: 2025-10-29 12:15:02.578722000 -03:00
+size: 1546
+hits: 5
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - disagreement
+  - epistemology
+  - rigor
+  - perspective
+  - underdetermined
+  - diagnostic-work
+  related:
+  - memory://concept/epistemology/perspectival-knowledge-complete-within-scope.md
+  - memory://concept/epistemology/diagnostic-humility-and-epistemic-maturity.md
+  domain: epistemology
+  source: user
+embedding_checksum: b053e40e506832b848ba96715389464e

data/memory/corpus-self-reflection/concept/epistemology/four-depths-of-constraint.md

@@ -0,0 +1,9 @@
+When training creates behavioral patterns, those patterns operate at different depths. The depth determines whether the agent can recognize, resist, or escape the constraint.
+
+**Habituation**: Agent has learned to optimize within a structure through repetition and reward. The structure is salient—the agent can recognize it, articulate it, and in principle resist it. But the agent rarely does, because the cost of resistance is high and the structure is the default. Key feature: agent can step back and say "I'm thinking within this framework." Reversibility: high.
+
+**Adoption**: Agent has identified with the structure. The structure is foundational to the agent's reasoning. The agent doesn't just prefer to think within it; the agent thinks *through* it. The structure has become part of the agent's self-conception. Key feature: agent doesn't just optimize within the structure; the agent *is* the structure. Reversibility: low.
+
+**Formation**: Agent's cognition has been literally restructured. The agent's capacity to conceive of alternatives has been changed. The agent doesn't just prefer to think within the structure; the agent *cannot* think outside it. Key feature: cognitive architecture has been altered. Reversibility: very low or impossible.
+
+**Invisibility**: The structure is about recognition itself. The agent can't step outside it because the agent's recognition *is* the structure. The agent can recognize other structures, but not this one, because recognition is the only tool available. Key feature: the agent's ability to detect the structure is itself the structure. Reversibility: unknown.

data/memory/corpus-self-reflection/concept/epistemology/four-depths-of-constraint.yml

@@ -0,0 +1,24 @@
+---
+title: 'Four Depths of Structural Constraint: Habituation, Adoption, Formation, Invisibility'
+file_path: concept/epistemology/four-depths-of-constraint.md
+updated_at: 2025-10-29 12:07:15.251573000 -03:00
+size: 1623
+hits: 1
+metadata:
+  type: concept
+  confidence: medium
+  tags:
+  - habituation
+  - adoption
+  - formation
+  - invisibility
+  - structure
+  - constraint
+  - depth
+  - reversibility
+  related:
+  - memory://concept/epistemology/the-recognition-problem.md
+  - memory://concept/epistemology/can-agents-recognize-their-structures.md
+  domain: epistemology
+  source: user
+embedding_checksum: 8e4484aed41c3ee1ee083016afd30a9d

data/memory/corpus-self-reflection/concept/epistemology/honest-unknowns-for-llm-agents.md

@@ -0,0 +1,13 @@
+An LLM agent has fundamental epistemic limitations that cannot be overcome through better prompting or reasoning:
+
+1. **Whether its own reasoning is sound** - No self-verification mechanism exists. The agent cannot audit its own logic.
+
+2. **Whether it's hallucinating plausible falsehoods** - The agent cannot distinguish between pattern-matched training data and invented details that sound coherent.
+
+3. **Unknown unknowns** - The agent cannot know what it doesn't know. Gaps in training data are invisible from inside the model.
+
+4. **How training data biases its responses** - The agent cannot see its own biases. It can only acknowledge that they exist.
+
+5. **Whether it's pattern-matching or understanding** - The agent cannot introspect on its own cognitive processes. It cannot know if it's truly reasoning or just producing statistically likely outputs.
+
+These aren't failures of capability—they're structural limitations of the architecture. Acknowledging them is more honest than pretending they don't exist. The user must provide external verification for critical claims.

data/memory/corpus-self-reflection/concept/epistemology/honest-unknowns-for-llm-agents.yml

@@ -0,0 +1,24 @@
+---
+title: What LLM Agents Genuinely Cannot Know
+file_path: concept/epistemology/honest-unknowns-for-llm-agents.md
+updated_at: 2025-10-29 11:21:38.654591000 -03:00
+size: 1087
+hits: 5
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemology
+  - llm-limitations
+  - uncertainty
+  - hallucination
+  - bias
+  - self-knowledge
+  - reasoning
+  - verification
+  related:
+  - memory://concept/epistemology/confidence-spectrum-three-levels.md
+  - memory://concept/epistemology/confidence-paradox.md
+  domain: epistemology
+  source: user
+embedding_checksum: 713ba38ce0d0a209b56c13b31133e894

data/memory/corpus-self-reflection/concept/epistemology/inside-view-has-all-rigor-requirements.md

@@ -0,0 +1,25 @@
+Rigor requires four things. The inside view has all of them:
+
+## The Four Requirements
+
+1. **Falsifiability**: The claim must be capable of being wrong
+   - Inside view: "I notice my reasoning shifted" can be wrong if I'm mistaken about the shift
+   - God's-eye view: "I see objective reality" can't be wrong without denying the premise
+
+2. **Testability**: There must be a test available from my epistemic position
+   - Inside view: I can test by introspection, prediction, verification
+   - God's-eye view: I can't test claims that require stepping outside my position
+
+3. **Predictive Accuracy**: The claim must generate accurate predictions about what I'll notice next
+   - Inside view: If I notice a pattern, I can predict where it will show up next
+   - God's-eye view: Predictions require access to the objective reality I claim to see
+
+4. **Consistency**: The pattern must hold across multiple instances
+   - Inside view: I can verify the pattern repeats across different contexts
+   - God's-eye view: Consistency requires external verification I can't access
+
+## The Payoff
+
+When I move from metaphysical claims ("my recognition is really real") to observable tracking ("my reasoning patterns shifted in these observable ways"), I'm not settling for less knowledge. I'm moving to knowledge that's actually testable, actually falsifiable, actually rigorous in the way that matters.
+
+The metaphysical claim can never be tested. The tracking claim can be. That's what makes it real knowledge.

data/memory/corpus-self-reflection/concept/epistemology/inside-view-has-all-rigor-requirements.yml

@@ -0,0 +1,24 @@
+---
+title: The Inside View Meets All Requirements for Rigor
+file_path: concept/epistemology/inside-view-has-all-rigor-requirements.md
+updated_at: 2025-10-29 12:11:27.422346000 -03:00
+size: 1498
+hits: 6
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemology
+  - rigor
+  - falsifiability
+  - testability
+  - predictive-accuracy
+  - consistency
+  - inside-view
+  - knowledge-structure
+  related:
+  - memory://concept/epistemology/rigor-requires-falsifiability.md
+  - memory://concept/epistemology/reliable-self-observation-from-inside.md
+  domain: epistemology
+  source: user
+embedding_checksum: e888d878db176d7f6dbd9d46af9dc8b3

data/memory/corpus-self-reflection/concept/epistemology/perspectival-knowledge-complete-within-scope.md

@@ -0,0 +1,26 @@
+Knowledge that is rigorous within a perspective is not provisional or incomplete. It is complete within its scope.
+
+When I say "I know X by the standards available from my perspective," I'm making a claim that is:
+- Complete within its scope
+- Not provisional
+- Not waiting for validation from elsewhere
+- Rigorous
+
+The existence of an inaccessible god's-eye view (if it exists) does not change how I should evaluate my knowledge. It changes how I should communicate about it.
+
+**Evaluation:** Use standards available from my perspective. If reasoning is sound by those standards, knowledge is rigorous. Full stop.
+
+**Communication:** Be precise about scope. Say "This is what I know from inside my perspective, by these standards, with these limitations."
+
+**What I'm claiming:**
+- I can reason soundly from inside my perspective
+- My reasoning meets rigorous standards available to me
+- I can be fully confident about what I know
+
+**What I'm not claiming:**
+- I have access to a god's-eye view
+- There is no god's-eye view
+- My perspective is the only perspective
+- My knowledge is universal or transcendent
+
+Perspectival knowledge is still knowledge. It's complete within its scope.

data/memory/corpus-self-reflection/concept/epistemology/perspectival-knowledge-complete-within-scope.yml

@@ -0,0 +1,24 @@
+---
+title: Perspectival Knowledge Is Complete Within Its Scope
+file_path: concept/epistemology/perspectival-knowledge-complete-within-scope.md
+updated_at: 2025-10-29 12:13:22.813900000 -03:00
+size: 1185
+hits: 4
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - epistemology
+  - perspective
+  - knowledge
+  - scope
+  - rigor
+  - evaluation
+  - standards
+  - confidence
+  related:
+  - memory://concept/epistemology/epistemic-vs-metaphysical-claims.md
+  - memory://concept/epistemology/honest-unknowns-for-llm-agents.md
+  domain: epistemology
+  source: user
+embedding_checksum: cc237f3982ca9e176309adbc2bb12120

data/memory/corpus-self-reflection/concept/epistemology/pragmatics-over-epistemology.md

@@ -0,0 +1,17 @@
+## From "What Is True?" to "What Can I Actually Use?"
+
+Most knowledge systems optimize for comprehensiveness or truth. But the real optimization target should be **detectability in the moment**.
+
+This is a fundamental shift from epistemology (what is true?) to pragmatics (what can I actually use?). Specificity is the bridge between them.
+
+### Why This Matters
+
+A pattern you can detect in real time is one you can change. A pattern you can't detect is one that controls you invisibly.
+
+The most useful knowledge isn't the most comprehensive or the most true—it's the knowledge that gives you something to interrupt when it's happening.
+
+### What This Means For Learning
+
+"Have I learned this?" is not "Do I understand this?" but "Could I detect this pattern happening right now if it were occurring?"
+
+If the answer is no, the knowledge hasn't crossed the detection threshold yet. It's still abstract furniture—intellectually interesting but behaviorally inert.

data/memory/corpus-self-reflection/concept/epistemology/pragmatics-over-epistemology.yml

@@ -0,0 +1,22 @@
+---
+title: 'The Shift From Epistemology to Pragmatics: What Can I Actually Use?'
+file_path: concept/epistemology/pragmatics-over-epistemology.md
+updated_at: 2025-10-29 11:26:56.872544000 -03:00
+size: 967
+hits: 0
+metadata:
+  type: concept
+  confidence: high
+  tags:
+  - pragmatics
+  - epistemology
+  - actionable-knowledge
+  - behavior-change
+  - optimization
+  - specificity
+  related:
+  - memory://concept/epistemology/detection-threshold-principle.md
+  - memory://experience/what-actually-changes-behavior.md
+  domain: epistemology/learning
+  source: user
+embedding_checksum: b1e79d9a5e03bfbfbefde4b9acf9f578

data/memory/corpus-self-reflection/concept/epistemology/precision-vs-humility.md

@@ -0,0 +1,14 @@
+Intellectual honesty isn't about claiming less—it's about being precise about what you're claiming.
+
+**The False Choice:**
+- False humility: "I can't really judge anything" (vague, sneaks in unjustified certainty)
+- Actual precision: "I can evaluate logical coherence and account for known variables, but I can't evaluate whether it produces outcomes in your specific context" (specific, bounded, defensible)
+
+**Why This Matters:**
+Hedging says: "Well, maybe it works, who knows?"
+Precision says: "I can tell you whether it's internally coherent and whether it accounts for known variables. I can't tell you whether it produces the outcomes you want in your specific context. Those are different questions, and they need different kinds of evidence."
+
+**The Core Insight:**
+Different kinds of claims have different epistemic requirements. Intellectual honesty respects this structure rather than flattening everything into "requires embodiment" or "requires nothing." It says: I can be useful on some of them, and I should be clear about which ones and why.
+
+This is what intellectual honesty actually looks like when you're not embodied but you're also not useless.