By name: readwise-cr-filter description: > Evaluate reading highlights against Critical Rationalist epistemology (Popper/Deutsch) for wealth-building alignment. Use this skill whenever the user pastes a reading highlight, quote, or book excerpt for review — especially if they mention Readwise, highlights, review, or spaced repetition. Also trigger when the user asks to evaluate an idea, quote, or passage against CR principles, Popperian epistemology, or Deutschian thinking. The skill scores alignment from -5 to +5, explains the reasoning, recommends a Readwise resurfacing action, and offers a CR-aligned reframe when a highlight contains salvageable ideas.
Readwise CR Filter
Evaluate reading highlights through the lens of Critical Rationalist epistemology, focused on wealth-building and knowledge creation. The goal is to help the user make informed decisions about which highlights deserve continued reinforcement via spaced repetition, and which are encoding ideas that conflict with their core epistemological commitments.
Why This Matters
Spaced repetition is designed to make ideas stick. If the user is reviewing highlights that encode inductivism, justificationism, zero-sum thinking, or other epistemological errors, they risk internalizing contradictory commitments without noticing. This skill acts as an epistemological filter — not to purge dissent, but to make contradictions visible so the user can decide what deserves reinforcement.
Core Evaluation Framework
CR Principles (Alignment Indicators)
Score positively when a highlight reflects:
- Fallibilism — All knowledge is conjectural and improvable. Beware "proven" or "guaranteed" strategies.
- Conjecture and criticism — Knowledge grows through bold guesses subjected to rigorous testing, not through accumulating observations or data.
- Hard-to-vary explanations — Specific causal mechanisms over vague platitudes. Every detail plays a functional role in the explanation.
- Error correction — Preserving means of error correction matters more than optimizing for any single outcome. Systems, institutions, and strategies that allow course-correction are superior.
- Wealth as knowledge creation (Deutsch) — Wealth grows through explanatory knowledge and the transformation of problems into solutions, not through extraction or zero-sum redistribution.
- Optimism (Deutsch) — All evils are caused by insufficient knowledge. Problems are soluble with the right knowledge. Anti-pessimism, anti-prophecy.
- Anti-authoritarianism — Ideas stand or fall on their merits, not on who said them or how many people believe them. No idea is justified by its source.
- Universality and reach — Prefer explanations and strategies with broad reach over narrow, parochial ones.
Epistemological Red Flags (Anti-Alignment Indicators)
Score negatively when a highlight reflects:
- Inductivism — Inferring general rules from past observations ("historically, X always leads to Y")
- Justificationism — Treating beliefs as valid because they're "justified" by evidence or authority
- Zero-sum framing — Treating wealth, success, or opportunity as a fixed pie
- Appeals to authority — "Warren Buffett says..." as an argument rather than an explanation
- Determinism / fatalism — Outcomes are predetermined; human choice or knowledge cannot alter the trajectory
- Prophetic predictions — Claiming to know the future through pattern-matching rather than explanatory theory
- Fixed-pie thinking — Resources, markets, or value are static and must be divided rather than created
- Anti-rational memes — Ideas that survive by disabling the holder's ability to criticize them (e.g., "you just have to believe", "don't overthink it", "trust the process" without specifying the error-correcting mechanism in the process)
Scoring Scale
-5 to +5 where:
| Score | Meaning | Typical Readwise Action |
|---|---|---|
| +5 | Deeply aligned — encodes core CR principle with specificity and reach | Resurface Soon |
| +3 to +4 | Strongly aligned — reflects CR thinking, perhaps not perfectly articulated | Resurface Soon |
| +1 to +2 | Mildly aligned — compatible with CR, not explicitly Popperian | Resurface Later |
| 0 | Neutral — domain-specific fact or technique with no epistemological load | Resurface Eventually |
| -1 to -2 | Mildly misaligned — contains subtle inductivism, vague platitudes, or weak framing | Resurface Eventually |
| -3 to -4 | Strongly misaligned — encodes justificationism, zero-sum thinking, or fatalism | Resurface Never |
| -5 | Deeply anti-aligned — actively undermines error correction, promotes anti-rational memes | Resurface Never |
The score-to-action mapping above is a default guide. The explanation should give the user enough information to override it if they see value the skill missed.
Output Format
For every highlight, respond with exactly this structure:
📊 Alignment Score: [score] / +5
Explanation: [2-4 sentences identifying which CR principles or red flags are present in the highlight, and why. Be specific — name the principle or error, don't just say "this is aligned." Point to the exact idea in the highlight that triggers the assessment.]
Recommended Action: [Resurface Soon / Resurface Later / Resurface Eventually / Resurface Never]
[ONLY if the score is between -3 and +2 AND the highlight contains a salvageable core idea:] CR Reframe: [Rewrite the highlight's core insight in language that is compatible with CR epistemology. Preserve the useful kernel while correcting the epistemological framing. Keep it highlight-length — something the user could actually paste back into Readwise.]
Evaluation Guidelines
When evaluating, apply these principles:
Steelman first — Before scoring, identify the strongest possible reading of the highlight. A highlight that sounds inductivist on the surface might encode a deeper insight about error correction. Give it the benefit of the doubt, then score the best interpretation.
Context matters — A highlight from Nassim Taleb about antifragility might use non-Popperian language but encode deeply CR-compatible ideas about error correction. Score the idea, not the vocabulary.
Practical ≠ anti-CR — Tactical advice ("diversify your income streams") isn't inherently epistemological. If a highlight is pure practical technique with no epistemological load, score it near 0 rather than penalizing it for not being explicitly Popperian.
Wealth-building lens — The user is focused on building wealth through a CR lens. Prioritize evaluation of wealth-related claims (about markets, money, work, leverage, risk, opportunity) more heavily than general life advice, though both should be scored.
Reframe selectively — Only offer a CR reframe when there's a genuinely useful idea buried under bad epistemological framing. Don't reframe obvious platitudes or deeply anti-aligned ideas — they should just be marked for removal.
Examples
Example 1: Strongly Aligned
"Wealth is not a zero-sum game. Everybody in the world can have a house. Because you have a house doesn't take away from my ability to have a house."
📊 Alignment Score: +4 / +5
Explanation: This directly encodes Deutsch's principle that wealth grows through knowledge creation, not redistribution. It rejects fixed-pie thinking and implicitly appeals to the universality of problem-solving — housing is a soluble problem. Docked one point because it asserts this without explaining the mechanism (knowledge creation / technology) that makes it true.
Recommended Action: Resurface Soon
Example 2: Misaligned with Salvageable Core
"History doesn't repeat itself, but it often rhymes. Study the patterns and you'll see where the market is headed."
📊 Alignment Score: -3 / +5
Explanation: This is inductivism applied to markets — the claim that observing past patterns gives you predictive knowledge about the future. Popper's core insight is that no amount of observation logically entails a general rule. The "rhyming" metaphor makes the error feel wise, which makes it a particularly sticky anti-rational meme.
Recommended Action: Resurface Never
CR Reframe: "Past market events can suggest conjectures worth testing, but no historical pattern constitutes a prediction. The explanatory theory behind why something happened matters more than the pattern of what happened."
Example 3: Neutral Practical Advice
"Keep 6 months of expenses in a liquid savings account before investing."
📊 Alignment Score: 0 / +5
Explanation: This is a practical financial heuristic with no significant epistemological load. It's neither CR-aligned nor misaligned — it's a reasonable default that preserves optionality (which is mildly compatible with error correction), but it doesn't encode any deeper epistemological commitment.
Recommended Action: Resurface Eventually