The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Allocate attention to agents based on lifecycle stage rather than salience, because deployment-stage agents need calibration attention while mature agents need minimal monitoring.
Recognize that expert-stage agents feel boring precisely because automaticity signals mastery—boredom with an agent is often evidence of successful maturation, not need for retirement.
Identify your biological prime time through multi-day tracking of hourly energy, focus, and motivation ratings rather than relying on self-perception, because subjective beliefs about peak performance times are systematically unreliable.
Retired agents compost into design knowledge for future agents—document what was learned, why it was retired, and what conditions made it obsolete, so nothing is lost.
Active unlearning (deliberate identification and retirement of obsolete knowledge) differs from passive forgetting and requires explicit lifecycle management to execute cleanly.
Insert an evaluation pause between receiving external input and adopting a conclusion, using that gap to explicitly assess evidence and reasoning against your own knowledge rather than defaulting to source credibility.
Build self-efficacy for independent judgment through accumulated mastery experiences—small decisions where you evaluate evidence yourself and observe outcomes—rather than through intellectual agreement with the concept of self-authority.
Deliberately separate the content of a recommendation from characteristics of its source by asking whether you would find the recommendation compelling if it came from a low-status source, exposing when you are responding to peripheral cues rather than substantive merit.
Before adopting an external recommendation, apply an accountability check by asking whether you would own the decision's consequences as your own or defer blame to the recommender, using the answer to diagnose whether you have retained or surrendered authority.
Recognize authority transfer through its felt signature—dissolving objections, premature agreement, bypassed evidence evaluation, or diffused responsibility—and use these somatic markers as triggers for metacognitive intervention before compliance completes.
Deliberately introduce processing friction when consulting AI by asking the same critical questions you would ask a junior colleague—what's the source, what assumptions underlie this, what's the strongest counterargument—rather than allowing fluency and confidence to bypass evaluation.
Seek authentic disagreement from people who genuinely hold contrary views rather than assigning devil's advocate roles, because only genuine dissent triggers the broader cognitive processing that improves judgment quality.
Break unanimity in conformity situations by being the first to dissent, because a single visible example of non-conformity reduces perceived social cost for everyone observing.
Block your measured peak attention hours on your calendar as recurring non-negotiable focus time and assign your most cognitively demanding tasks to those blocks, because cognitive performance varies 7-40% across the day and high-stakes decisions made during depletion are measurably worse.
Start reclaiming authority in domains with lowest combined consequence severity and social friction to build mastery experiences before tackling high-stakes domains.
Take clear I-positions rather than we-positions or appeals to external authority to make your thinking visible and updatable.
Maintain emotional presence and contact with relational partners during disagreement rather than withdrawing, distinguishing differentiation from emotional cutoff.
Distinguish being influenced (changing mind due to persuasive argument) from being controlled (changing position due to emotional discomfort with disapproval).
When you possess domain-specific expertise relevant to a decision, voice your dissenting assessment even when it conflicts with hierarchical authority, because withholding situated knowledge creates epistemic fragility and constitutes professional negligence.
Frame professional dissent as questions rather than assertions when building credibility, because questions activate information-sharing while reducing social cost and allowing decision-makers to reach conclusions themselves.
Build authority to dissent through demonstrated competence and calibrated prediction tracking, because self-authority without proven expertise is indistinguishable from arrogance and carries no credible weight.
Replace algorithmically curated feeds with deliberately selected information sources, because algorithmic curation optimizes for engagement rather than accuracy and systematically filters input to maximize emotional reactivity.
Audit your beliefs for algorithmic origins by tracing each position to its informational source, because beliefs acquired through engineered exposure rather than deliberate inquiry have not been subjected to sovereign epistemic standards.
When using AI systems, maintain explicit separation between input generation (AI-assisted) and judgment synthesis (human-retained), because outsourcing the integration function constitutes epistemic abdication regardless of input quality.