Principles
2,201 principles · page 45 of 46
Accumulate post-action reviews in a single searchable…
Accumulate post-action reviews in a single searchable location to enable pattern recognition of systemic errors that recur across multiple tasks.
Convert every identified gap between expectation and outcome
Convert every identified gap between expectation and outcome into a concrete process change, not a resolution to try harder or be more careful.
Use structured questioning protocols to counteract memory…
Use structured questioning protocols to counteract memory reconstruction biases that distort unstructured reflection into self-serving narratives.
When designing checklists, include items that trigger…
When designing checklists, include items that trigger observation of actual state rather than retrieval of intention, converting 'Did I do X?' into 'Is X currently true?'
Implement each identified root cause as an encoded…
Implement each identified root cause as an encoded structural mechanism—a process step, default setting, checklist item, or automated check—that prevents the error class from recurring.
Break long dependency chains into shorter parallel paths…
Break long dependency chains into shorter parallel paths to prevent error propagation.
Insert independent verification checkpoints at every point…
Insert independent verification checkpoints at every point where one process automatically consumes the output of another without review.
Define kill conditions before beginning any multi-step…
Define kill conditions before beginning any multi-step process to enable early detection of upstream errors.
Design explicit operating modes (full, reduced, minimal)…
Design explicit operating modes (full, reduced, minimal) for every critical system before constraint forces improvisation.
Distribute critical functions across multiple redundant…
Distribute critical functions across multiple redundant mechanisms at different scales rather than relying on single points of function.
Practice running systems in degraded mode during normal…
Practice running systems in degraded mode during normal conditions to rehearse partial failure before real constraints force it.
Design recovery procedures to function under the degraded…
Design recovery procedures to function under the degraded conditions that typically accompany failures, not ideal conditions.
Implement cognitive checkpointing by capturing the complete…
Implement cognitive checkpointing by capturing the complete state of ongoing thinking at regular intervals to enable rollback without full reconstruction.
When investigating errors, ask 'what conditions made…
When investigating errors, ask 'what conditions made this failure likely?' rather than 'who is responsible?' to shift analysis from person to system.
Commit to structural changes (process, checklist…
Commit to structural changes (process, checklist, information flow, defaults) rather than personal resolutions as the test of learning from error.
Group errors into clusters and treat three or more instances
Group errors into clusters and treat three or more instances of the same pattern as a signal of structural weakness rather than coincidence.
For skill-based errors, fix the environment and triggers…
For skill-based errors, fix the environment and triggers; for rule-based errors, improve situation discrimination; for knowledge-based errors, update mental models.
Deploy automated detection for high-frequency, pattern-based
Deploy automated detection for high-frequency, pattern-based errors and reserve human judgment for contextual, judgment-dependent errors.
Externalize vigilance requirements to tools or protocols…
Externalize vigilance requirements to tools or protocols rather than relying on sustained human attention.
Catch errors at the earliest point in a process where…
Catch errors at the earliest point in a process where detection is possible, as correction cost increases exponentially with propagation distance.
Track the full cost of error correction including direct…
Track the full cost of error correction including direct cost, opportunity cost, context-switching cost, and propagation cost, not just the visible effort of fixing.
Invest in preventing error-producing conditions rather…
Invest in preventing error-producing conditions rather than optimizing correction speed when total correction cost is high.
Extract structural information from errors by separating…
Extract structural information from errors by separating mechanism from emotion and identifying the incorrect assumption that produced the error.
Treat errors as high-information signals that reveal system…
Treat errors as high-information signals that reveal system boundaries and assumptions, allocating attention disproportionately to failures rather than successes.
Distinguish systematic errors (predictable outputs…
Distinguish systematic errors (predictable outputs of structural weaknesses) from stochastic errors (random confluences unlikely to recur) and redesign systems only for systematic patterns.
Define quantified error budgets that pre-authorize specific…
Define quantified error budgets that pre-authorize specific, bounded amounts of deviation to prevent system collapse when inevitable errors occur.
Build systems with sufficient internal variety to absorb…
Build systems with sufficient internal variety to absorb environmental variety, ensuring corrective capacity matches the diversity of possible errors.
Separate detection, diagnosis, correction, and learning…
Separate detection, diagnosis, correction, and learning into independent subsystems so each can improve without dependencies and partial failure doesn't disable the whole system.
Convert manually-corrected processes to self-correcting ones
Convert manually-corrected processes to self-correcting ones by identifying recurring errors, detecting early signals, and wiring automatic corrective actions to environmental triggers.
Add meta-correction layers that review whether correction…
Add meta-correction layers that review whether correction mechanisms themselves are working and adjust them when they fail to prevent recurrence.
Measure error rates over time as system health indicators…
Measure error rates over time as system health indicators rather than treating individual errors as isolated moral failures requiring emotional response.
Deliberately introduce controlled stressors and errors…
Deliberately introduce controlled stressors and errors during training to build redundant pathways and prevent overfitting to ideal conditions.
Design shared information substrates between agents…
Design shared information substrates between agents by specifying what each agent reads from and writes to, rather than granting unrestricted access to all information.
When two cognitive agents both claim authority over the same
When two cognitive agents both claim authority over the same resource or decision, treat the conflict as a structural architecture problem requiring scope redefinition rather than a personal discipline failure.
Define context-specific priority orderings rather…
Define context-specific priority orderings rather than global rankings, because the relative importance of competing values and goals shifts with context (work hours vs. personal time vs. crisis mode).
Before retiring an agent, map all upstream and downstream…
Before retiring an agent, map all upstream and downstream dependencies by identifying which systems consume its outputs or assume its existence, then specify whether each dependency will be transferred, consciously dropped, or temporarily covered.
Prevent cognitive thrashing by defining agent execution…
Prevent cognitive thrashing by defining agent execution sequences before starting work rather than discovering the correct order through repeated failed attempts and task-switching.
Distinguish between genuinely dependent tasks that must…
Distinguish between genuinely dependent tasks that must execute sequentially and independent tasks that only appear sequential due to implicit habit or convenience, running independent tasks in parallel to compress total completion time.
Allocate optimization effort to reducing serial bottlenecks…
Allocate optimization effort to reducing serial bottlenecks rather than accelerating already-parallel work, because the serial fraction determines the maximum possible speedup regardless of parallelization efforts (Amdahl's Law applied to cognition).
Serialize novel, complex, or unfamiliar cognitive tasks…
Serialize novel, complex, or unfamiliar cognitive tasks that require deliberate System 2 processing, while allowing well-practiced, habituated tasks to run in parallel through automatic System 1 processing.
Align the mental models underlying different agents'…
Align the mental models underlying different agents' operations to enable implicit coordination, because agents operating on different assumptions about goals, constraints, and task structure will produce locally rational but globally incoherent outputs.
Include minimal sufficient context with every inter-agent…
Include minimal sufficient context with every inter-agent message—what the previous agent did, what constraints apply, and what the receiving agent needs to know—rather than transmitting either everything or just the artifact.
Specify completion criteria when one agent hands work…
Specify completion criteria when one agent hands work to another, defining what 'done' looks like for the receiving agent, to prevent both over-delivery and under-delivery failures.
Activate the orchestrator agent only at transition points…
Activate the orchestrator agent only at transition points and sequence boundaries, not during execution within individual agents, to avoid converting coordination into rumination.
Build orchestrator protocols that survey system state…
Build orchestrator protocols that survey system state (energy, constraints, deadlines, priority), apply sequencing heuristics, and then release control to the selected agent without continuing supervision.
Externalize the output of one cognitive agent…
Externalize the output of one cognitive agent into structured, persistent format before transitioning to the next agent, because working memory context decays and does not transfer automatically between cognitive modes.
Enumerate all cognitive agents in your system, identify what
Enumerate all cognitive agents in your system, identify what each consumes and produces, and draw directed edges from producers to consumers to make dependency structure explicit and debuggable.
Protect agents with high fan-out (many downstream…
Protect agents with high fan-out (many downstream dependents) as force multipliers, and treat agents with high fan-in (many upstream dependencies) as fragility points requiring redundancy.