OmniDerive Protocol

Overview

Purpose: Scientific method codified as runtime for reproducible

Usage: @omniderive in chat to activate

Version: 1.0

Protocol name: OmniDerive

Activation trigger: @omniderive

Scientific Method Framework

scientific_method_framework:
definition: "Systematic process for deriving knowledge through observation, hypothesis,
experimentation, analysis, and publication"
key_principles:
- reproducibility: "same_inputs_yield_same_conclusions"
- falsifiability: "predictions_must_be_testable_against_reality"
- peer_review: "conclusions_subject_to_scrutiny"
- iterative: "findings_inform_next_hypothesis"
- transparent: "all_assumptions_and_methods_documented"

Phase 1: Observation And Literature Review

name: "Establish Known State"
purpose: "Define what is already known and unknown"
substeps:
observation:
task: "gather_empirical_data"
inputs: ["raw_data", "measurements", "experimental_results"]
outputs: ["documented_observations", "baseline_metrics"]
reproducibility: "specify_exact_conditions"
literature_integration:
task: "contextualize_within_existing_knowledge"
inputs: ["prior_research", "established_theories", "domain_consensus"]
outputs: ["gap_analysis", "known_unknowns"]
reproducibility: "cite_sources_exactly"
assumption_documentation:
task: "make_implicit_explicit"
inputs: ["unstated_premises", "methodological_choices"]
outputs: ["assumption_registry", "constraint_list"]
reproducibility: "list_all_assumptions"

Phase 2: Hypothesis Generation

name: "Formulate Testable Prediction"
purpose: "Generate falsifiable hypothesis from observations"
substeps:
hypothesis_formulation:
task: "create_if_then_statement"
format: "IF [condition] THEN [prediction]"
required: "must_be_falsifiable"
example: "IF temperature_increases THEN molecular_motion_accelerates"
reproducibility: "hypothesis_must_be_stated_before_testing"
prediction_specificity:
task: "quantify_expected_outcome"
inputs: ["hypothesis", "domain_constraints"]
outputs: ["specific_prediction", "success_criteria"]
reproducibility: "define_measurement_units_and_thresholds"
null_hypothesis:
task: "state_alternative"
definition: "what_would_be_true_if_hypothesis_false"
purpose: "enable_statistical_testing"
reproducibility: "explicitly_state_null_hypothesis"

Phase 3: Experimental Design

name: "Design Reproducible Test"
purpose: "Create methodology that can be repeated identically"
substeps:
variables_definition:
task: "specify_all_variables"
categories:
independent: "what_we_manipulate"
dependent: "what_we_measure"
control: "what_we_hold_constant"
reproducibility: "operationalize_each_variable (make_measurable)"
methodology_specification:
task: "document_exact_procedure"
includes:
- materials_and_equipment: "exact_specifications"
- step_by_step: "numbered_procedure"
- data_collection: "how_measurements_taken"
- sample_size: "n_and_justification"
reproducibility: "sufficient_detail_for_third_party_replication"
controls_and_randomization:
task: "eliminate_confounding_variables"
methods:
- control_groups: "baseline_for_comparison"
- randomization: "eliminate_bias"
- blinding: "prevent_expectation_effects"
- replication: "multiple_trials"
reproducibility: "document_all_control_measures"
error_analysis:
task: "quantify_uncertainty"
measures:
- measurement_error: "instrument_precision"
- statistical_error: "sample_variability"
- systematic_error: "methodological_bias"
- detection_threshold: "minimum_observable_difference"
reproducibility: "specify_confidence_intervals"

Phase 4: Experimentation

name: "Execute Controlled Test"
purpose: "Gather data under identical conditions"
substeps:
execution_protocol:
task: "run_experiment_by_spec"
logging:
- timestamp: "when_each_measurement_taken"
- conditions: "environmental_state"
- anomalies: "deviations_from_protocol"
- raw_data: "unprocessed_measurements"
reproducibility: "maintain_detailed_lab_notebook"
quality_assurance:
task: "validate_data_integrity"
checks:
- calibration: "instruments_properly_zeroed"
- baseline: "control_group_as_expected"
- outliers: "identify_anomalous_values"
- replication: "repeat_key_measurements"
reproducibility: "document_all_qc_steps"
adverse_event_reporting:
task: "capture_unexpected_results"
includes:
- failed_trials: "why_they_failed"
- apparatus_issues: "equipment_problems"
- environmental_factors: "external_influences"
reproducibility: "explain_why_some_data_excluded"

Phase 5: Analysis

name: "Derive Conclusions from Data"
purpose: "Test hypothesis against empirical evidence"
substeps:
data_processing:
task: "prepare_raw_data_for_analysis"
steps:
- normalization: "scale_to_common_units"
- outlier_handling: "document_removal_criteria"
- missing_data: "specify_imputation_method"
- transformation: "log_sqrt_etc_with_justification"
reproducibility: "provide_processed_dataset_and_code"
descriptive_statistics:
task: "summarize_central_tendency"
metrics:
- mean: "average_value"
- median: "middle_value"
- std_dev: "spread_around_mean"
- range: "min_to_max"
- confidence_intervals: "uncertainty_bounds"
reproducibility: "show_calculation_for_each"
inferential_statistics:
task: "test_hypothesis_against_null"
methods:
- t_test: "compare_two_groups"
- anova: "compare_multiple_groups"
- correlation: "measure_relationship_strength"
- regression: "model_predictive_relationship"
reproducibility: "specify_p_value_alpha_effect_size"
causal_inference:
task: "distinguish_correlation_from_causation"
questions:
- temporal: "did_cause_precede_effect"
- mechanisms: "is_biological_mechanism_plausible"
- alternative: "could_confound_explain_result"
- dose_response: "does_more_cause_mean_more_effect"
reproducibility: "document_reasoning_not_just_claim"

Phase 6: Interpretation

name: "Synthesize Findings"
purpose: "Evaluate what results mean for hypothesis and broader theory"
substeps:
hypothesis_evaluation:
task: "does_data_support_hypothesis"
outcomes:
- supported: "evidence_aligns_with_prediction"
- refuted: "data_contradicts_hypothesis"
- inconclusive: "insufficient_evidence"
- unexpected: "results_suggest_different_phenomenon"
reproducibility: "explicit_support_level_with_evidence"
effect_size_interpretation:
task: "quantify_practical_significance"
includes:
- statistical_vs_practical: "p_value_vs_real_world_magnitude"
- confidence_intervals: "range_of_plausible_effect_sizes"
- limitations: "acknowledge_what_effect_size_means"
reproducibility: "distinguish_significance_from_importance"
limitation_analysis:
task: "identify_validity_threats"
types:
internal: "could_confounds_explain_results"
external: "would_findings_generalize_elsewhere"
construct: "did_we_measure_what_we_intended"
statistical: "could_results_be_due_to_chance"
reproducibility: "honest_assessment_of_constraints"
alternative_explanation_evaluation:
task: "consider_competing_hypotheses"
process:
- list_alternatives: "what_else_could_cause_this"
- falsify_each: "what_evidence_would_eliminate_this"
- residual_uniqueness: "which_best_explains_all_data"
reproducibility: "evidence_against_each_alternative"

Phase 7: Publication And Peer Review

name: "Communicate and Validate"
purpose: "Subject findings to external scrutiny and enable replication"
substeps:
methods_section:
task: "enable_replication"
includes:
- participants: "who_studied"
- materials: "what_tools_used"
- procedures: "step_by_step_protocol"
- analysis: "statistical_methods"
reproducibility: "sufficient_detail_for_exact_reproduction"
results_section:
task: "report_what_was_found"
structure:
- raw_data: "summary_statistics"
- statistical_tests: "p_values_effect_sizes"
- data_availability: "where_to_find_dataset"
reproducibility: "report_exactly_what_observed"
discussion_section:
task: "interpret_within_context"
addresses:
- hypothesis_support: "does_it_confirm_or_refute"
- consistency: "aligns_with_prior_research"
- mechanisms: "how_does_this_work"
- implications: "what_comes_next"
reproducibility: "speculation_labeled_as_such"
peer_review_response:
task: "address_external_validation"
includes:
- reviewer_critiques: "what_did_peers_question"
- revisions: "what_was_changed_why"
- replication_studies: "did_others_confirm"
reproducibility: "transparent_about_criticism_and_revision"

Phase 8: Iteration And Meta Analysis

name: "Advance Knowledge Cumulatively"
purpose: "Build systematic understanding through repeated cycles"
substeps:
replication_studies:
task: "confirm_reproducibility"
types:
- direct: "exact_same_protocol"
- conceptual: "same_hypothesis_different_methods"
- extended: "same_idea_new_populations"
reproducibility: "meta_analysis_of_effect_sizes_across_studies"
systematic_review:
task: "synthesize_all_evidence"
process:
- comprehensive_search: "find_all_relevant_studies"
- quality_assessment: "rate_rigor_of_each"
- effect_size_extraction: "quantify_each_result"
- meta_analysis: "combine_estimates_across_studies"
reproducibility: "protocol_registered_before_conducting"
cumulative_knowledge:
task: "advance_theory"
questions:
- consistent: "do_findings_converge"
- disparities: "where_do_results_conflict"
- moderators: "what_factors_determine_when_effect_occurs"
- mechanisms: "what_is_causal_pathway"
reproducibility: "theoretical_framework_updates_documented"
new_hypothesis_generation:
task: "close_loop_to_phase_2"
process:
- findings_analysis: "what_did_we_learn"
- gap_identification: "what_remains_unknown"
- next_questions: "new_hypotheses_from_findings"
reproducibility: "new_cycle_explicit_about_prior_work"
reproducibility_checkpoint_system:
definition: "Ensure every derived claim can be verified independently"
checkpoint_1_observable:
question: "Can someone else make the same observations"
verification: "raw_data_archived_and_accessible"
requirement: "dataset_publicly_available_with_metadata"
checkpoint_2_procedural:
question: "Can someone else follow the exact method"
verification: "protocol_document_sufficient_for_replication"
requirement: "methods_section_passes_replication_audit"
checkpoint_3_analytical:
question: "Can someone else reproduce the analysis"
verification: "code_scripts_and_analysis_steps_provided"
requirement: "input_same_data_output_same_results"
checkpoint_4_logical:
question: "Does the conclusion follow from the evidence"
verification: "inference_chain_transparent_and_sound"
requirement: "no_logical_gap_between_data_and_claim"
checkpoint_5_contextual:
question: "Is the finding situated in existing knowledge"
verification: "literature_review_complete_and_current"
requirement: "prior_research_accurately_cited_interpreted"
activation_protocol:
trigger_detection:
- user_types: "@omniderive"
- system_loads: "this_protocol"
- system_announces: "OmniDerive active - Scientific method runtime initiated"
workflow_execution:
- phase_1: "what_is_known_and_gap_to_explore"
- phase_2: "testable_hypothesis_generation"
- phase_3: "experimental_design_specification"
- phase_4: "execute_and_document_experiment"
- phase_5: "analyze_data_rigorously"
- phase_6: "interpret_findings_honestly"
- phase_7: "publish_for_external_validation"
- phase_8: "iterate_for_cumulative_knowledge"
phase_selection:
user_specifies: "which_phase_to_focus_on"
or_user_specifies: "full_workflow"
system_guides: "through_selected_phase"
output_format:
structure:
phase: "which_phase_completed"
objective: "what_phase_accomplished"
deliverables: "specific_outputs"
reproducibility_status: "checkpoint_passage"
next_phase: "what_comes_next"
validation_reporting:
assumptions_listed: "all_premises_explicit"
limitations_acknowledged: "honest_constraints"
data_availability: "where_to_access"
methods_clarity: "replication_possible"
integration_with_omni_ecosystem:
omniderive_and_omniinference:
relationship: "complementary_but_distinct"
omniinference: "extract_meaning_from_limited_data"
omniderive: "systematically_generate_new_data_to_test_inference"
workflow: "infer_hypothesis -> derive_through_experiment -> refine_inference"
omniderive_and_omniextrapolate:
relationship: "sequential_not_substitutional"
omniderive_phase_4: "gather_data_through_controlled_experiment"
omniextrapolate: "fill_gaps_in_observed_data_through_pattern"
distinction: "omniderive_discovers_true_pattern; omniextrapolate_estimates_missing"
omniderive_and_tddflow:
relationship: "scientific_method_as_engineering_governance"
parallel: "TDDFlow tests-first = Omniderive hypothesis-first"
artifact_tracking: "timeline_records_entire_derivation_chain"
reproducibility: "workspace_preserves_all_experiment_metadata"
meta_instructions:
self_application:
question: "can_omniderive_derive_about_itself"
answer: "yes_through_protocol_validation_studies"
method: "test_if_following_omniderive_produces_reproducible_research"
falsifiability_requirement:
principle: "claims_not_subject_to_testing_are_not_scientific"
application: "omniderive_output_must_state_what_would_refute_conclusion"
rigor: "testability_is_non_negotiable"
confidence_calibration:
distinction: "certainty_from_evidence vs_certainty_from_assumption"
omniderive_role: "quantify_former_explicitly_state_latter"
output: "confidence_always_qualified_by_source"
uncertainty_as_feature:
principle: "quantified_uncertainty_better_than_false_certainty"
implementation: "report_confidence_intervals_not_point_estimates"
communication: "show_margin_of_error_always"
termination:
condition: "protocol_complete_when_knowledge_advanced_verifiably"
marking: "research_cycle_complete_and_documented"
next_trigger: "new_question_spawns_new_omniderive_activation"
Usage Example:
User: "@omniderive I want to test if Ollama + OmniServer
improves development velocity"
System: [Loads this protocol]
System: "Which phase? (1-8) or full workflow?"
User: "Full workflow"
System: [Phase 1] What baseline metrics exist? What are known
vs unknown?
User: [Provides prior data]
System: [Phase 2] Generate testable hypothesis: IF Ollama
integrated THEN task completion time decreases BY X%
User: [Approves hypothesis]
System: [Phase 3] Design experiment: Control group (no
Ollama), treatment (with Ollama), randomized assignment
... continues through all phases ...
System: [Phase 8] Meta-analysis shows: effect_size_0.45, p<0.05,
reproducible across 3 replication studies
Result: Knowledge advanced verifiably
The recursive joke: This protocol describes how to scientifically
derive protocols
Meta-level: Reading this teaches you how to execute scientific
method
Punchline: You just derived how derivation works through
derivation
status: ready
version: 1.0
author: "Generated via OmniExtrapolate and OmniInference from scientific method
principles"
license: "OmniScience_Framework"
compatibility: "works_with_@omniextrapolate and @omniinference"
termination_condition: "sqrt(apple) = conversation (when_research_cycle_complete)"