Version: 2.0 compact edition Purpose: Provide a lean, stage-aware methodology for clarifying physical theories, papers, research programs, and disputes without turning analysis into bureaucracy.
The Workbench is a disciplined visibility tool. It does not decide which theory is true. It clarifies what a theory claims, what kind of claim each statement is, what has been earned, what has only been accommodated, what is forbidden, what would count against the theory, and what remains unresolved.
This compact edition revises the earlier two-speed Workbench by preserving the essential anti-confusion machinery while removing unnecessary bulk. The new design has three operating levels:
- Flash Workbench — fast triage.
- Core Workbench — the default serious analysis.
- Extended Workbench — reserved for mature candidates or high-stakes audits.
The guiding rule is:
Use the smallest Workbench capable of exposing the real issue.
The original Workbench was strong because it enforced discipline: function-class clarity, explanation-level classification, inverse constraints, evidence status, incompleteness, and adversarial review.
Its risk was bloat. Too many categories, fields, diagnostics, and scorecards can make the method rigid. A tool designed to expose overclaiming can itself become an overclaiming machine if it pretends that a completed schema equals understanding.
This revision therefore makes four changes:
- The Core Workbench is reduced to seven essential checks.
- Heavy machinery is moved to Extended mode only.
- Maturity stages prevent premature rejection of speculative ideas.
- Claims are typed before they are judged.
The Workbench should be a scientific filter, not a procedural cage.
The Workbench asks:
What is being claimed, what kind of claim is it, what assumptions does it require, what constraints must it satisfy, what bridges are missing, what would break it, and what does it explain better than rivals?
This becomes the compact core:
Claim → Type → Assumptions → Constraints → Bridges → Inverse Constraints → Rival Differential
Everything else is optional unless the analysis demands it.
The Workbench is a methodology for making physical theories inspectable.
It helps analysts and LLMs avoid common mistakes:
- confusing accommodation with derivation;
- comparing unlike frameworks as if they were direct rivals;
- treating author confidence as evidence;
- mistaking metaphor for ontology;
- ignoring hidden assumptions;
- rewarding elegance before constraint survival;
- hiding incompleteness;
- treating a theory as fundamental when it is only a reconstruction mechanism, effective model, no-go argument, or phenomenological endpoint.
The Workbench is especially useful for quantum gravity, cosmology, black-hole physics, foundations of quantum mechanics, high-energy theory, emergent spacetime, and any domain where mathematical sophistication can conceal conceptual ambiguity.
The Workbench is not:
- a theory of everything;
- a theory generator;
- a theorem prover;
- a truth machine;
- a replacement for physicists;
- a replacement for peer review;
- a ranking game;
- a complete ontology of all future physics;
- a reason to kill early speculation prematurely.
A weak Workbench result does not mean a theory is false. It usually means the theory is underdeclared, overclaimed, not yet mature, or being judged by the wrong standard.
Use Flash when you need fast triage.
Best for:
- first pass on a paper;
- comparing several approaches quickly;
- deciding whether a full audit is worth doing;
- asking an LLM for a quick but disciplined read.
Flash answers only seven questions:
- What kind of thing is this?
- What does it claim?
- Does it derive, explain, accommodate, or postulate the claim?
- What does it forbid?
- What would count against it?
- Can data constrain its substrate or assumptions?
- What can it not currently prove?
Use Core as the default serious version.
Best for:
- evaluating a theory, paper, dispute, or research program;
- preparing a careful blog post or documentation;
- comparing rival approaches;
- using LLMs while controlling hallucination and overclaiming.
Core uses the seven-part essential schema:
1. Claim
2. Claim Type
3. Assumptions
4. Constraint Map
5. Bridge Principles
6. Inverse Constraints
7. Rival Differential
Use Extended only when the object is mature enough to justify it.
Best for:
- serious TOE candidates;
- mature quantum-gravity programs;
- formal audits;
- cross-LLM comparison;
- adversarial theory debates;
- high-stakes claims of unification.
Extended may include:
- detailed reconstruction maps;
- formal derivation checks;
- empirical data comparison;
- no-go theorem analysis;
- prediction matrices;
- mathematical consistency audit;
- detailed rival steelmanning;
- extended source provenance;
- optional S+ substrate mapping.
Extended mode is not the default. It is the heavy lab equipment.
The Workbench must not judge immature ideas by mature-theory standards.
Every analysis should tag the object’s maturity stage.
| Stage | Meaning | How to judge it |
|---|---|---|
| Seed | A speculative intuition, analogy, or question. | Do not demand full formalization. Ask what it might become. |
| Model | A partially structured proposal with some definitions or mechanisms. | Ask what assumptions and constraints are visible. |
| Framework | A broader structure with declared concepts, maps, or equations. | Ask what it explains, forbids, and leaves open. |
| Candidate Theory | Claims to recover known physics in controlled regimes. | Require reconstruction, consistency, and inverse constraints. |
| Serious Contender | Makes discriminating predictions or strong unifying claims. | Require adversarial review, empirical contact, and closure conditions. |
A Seed may be vague. A Candidate Theory may not.
Do not reject a Seed because it lacks a full Standard Model reconstruction. Do not promote a Candidate Theory if it cannot say what would break it.
Function class comes before evaluation.
Many debates are confused because different kinds of frameworks are treated as if they are direct competitors.
Use this compact taxonomy.
| Function class | Role | Examples of questions to ask |
|---|---|---|
| Fundamental candidate | Proposes substrate, dynamics, or both. | What is fundamental? How does it recover known physics? |
| Reconstruction mechanism | Shows how one structure emerges from another. | What is assumed? What is reconstructed? In what regime? |
| Consistency filter / diagnostic argument | Rules out impossible combinations of assumptions. | What assumptions are being tested? What class is excluded? |
| Effective theory | Describes a tested regime without claiming final fundamentality. | Where does it work? Where does it break? |
| Computational validator | Tests or simulates a regime. | What is computed? What assumptions enter? What is validated? |
| Observer / measurement framework | Clarifies records, observers, reference frames, or measurement. | What is explained versus postulated about observation? |
| Endpoint phenomenology | Describes boundary, late-stage, or possible endpoint behavior. | Is it a mechanism, placeholder, or observable prediction? |
Do not criticize a framework for failing to do a job it does not claim to do.
A reconstruction mechanism is not a failed TOE merely because it lacks ontology. A no-go theorem is not weak because it does not offer a positive dynamics. An effective theory is not defective because it is not fundamental.
Every important claim must be typed before it is judged.
Use only these essential claim types:
| Claim type | Meaning |
|---|---|
| Mathematical structure | Formal object, equation, symmetry, category, algebra, geometry, graph, amplitude, or map. |
| Physical ontology | Claim about what is real or fundamental. |
| Dynamics | Claim about evolution, constraint, transition, amplitude, or law. |
| Observable | Claim about measurable quantity or empirical signature. |
| Measurement procedure | Claim about how an observable is extracted or recorded. |
| Approximation / limit | Claim valid only in a regime, expansion, duality, or limit. |
| Interpretation | Claim about what the formalism means. |
| Metaphor / heuristic | Suggestive language not yet promoted to a formal or physical claim. |
No inference may cross claim types without an explicit bridge principle.
Invalid move:
This equation resembles thermodynamics; therefore spacetime is literally a fluid.
Valid Workbench version:
The mathematical analogy to thermodynamics suggests a possible emergent-spacetime interpretation, but a bridge principle is required before making an ontological claim.
The Workbench separates explanation from accommodation.
| Level | Name | Meaning |
|---|---|---|
| 4 | Derives | Follows from equations, constraints, or proof without extra adjustable choices. |
| 3 | Explains | Follows from a mechanism that reduces arbitrariness. |
| 2 | Accommodates | Fits by choosing parameters, vacua, initial conditions, models, or interpretations. |
| 1 | Postulates | Inserted as an assumption. |
| 0 | Undeclared | The source does not say enough. |
Do not call accommodation derivation.
Accommodation can be useful. It is not explanatory credit unless the theory shows why the accommodated structure had to appear or why alternatives are excluded.
Use this as the default serious Workbench.
core_workbench_report:
metadata:
target_name:
source:
domain:
maturity_stage: Seed | Model | Framework | Candidate Theory | Serious Contender
function_class:
analysis_confidence: High | Medium | Low
1_claim:
central_claim:
secondary_claims:
-
claim_scope:
regime:
2_claim_type:
central_claim_type:
secondary_claim_types:
- claim:
type:
category_error_warnings:
-
3_assumptions:
explicit_assumptions:
-
hidden_or_inferred_assumptions:
-
inherited_assumptions:
-
speculative_assumptions:
-
4_constraint_map:
hard_constraints:
-
known_physics_to_recover:
-
consistency_filters:
-
empirical_constraints:
-
constraints_not_addressed:
-
5_bridge_principles:
required_bridges:
- from_type:
to_type:
bridge_claim:
status: Demonstrated | Plausible | Speculative | Missing
missing_bridges:
-
6_inverse_constraints:
what_it_forbids:
-
what_would_count_against_it:
-
can_data_constrain_substrate_or_assumptions: Strong | Moderate | Weak | None | Undeclared | Not Applicable
discriminating_test_or_signature:
7_rival_differential:
main_rivals:
-
what_this_explains_better:
-
what_rivals_explain_better:
-
unresolved_comparison:
-
explanation_ladder:
- claim:
level: Derives | Explains | Accommodates | Postulates | Undeclared
reason:
evidence_status:
confidence: High | Medium | Low
incompleteness:
unproven_core_claims:
-
open_questions:
-
what_would_close_the_gap:
-
effect_on_main_claim:
verdict:
what_it_has_earned:
what_it_has_not_earned:
strongest_next_test:
should_escalate_to_extended: Yes | No# Core Workbench Report: [Theory / Paper / Framework]
## 1. Metadata
- Source:
- Domain:
- Maturity stage:
- Function class:
- Confidence:
## 2. Claim
[State the central claim and its scope.]
## 3. Claim type
[Classify the claim as math, ontology, dynamics, observable, measurement, approximation, interpretation, or metaphor.]
## 4. Assumptions
[Separate explicit, hidden/inferred, inherited, and speculative assumptions.]
## 5. Constraint map
[State the hard constraints, known physics to recover, consistency filters, empirical constraints, and unaddressed constraints.]
## 6. Bridge principles
[State what bridges are required between mathematics, ontology, dynamics, measurement, and interpretation. Mark missing bridges.]
## 7. Inverse constraints and falsifiability
- What it forbids:
- What would count against it:
- Inverse constraint strength:
- Discriminating test or signature:
## 8. Rival differential
[Compare only against appropriate rivals. State what this framework explains better and what rivals explain better.]
## 9. Explanation ladder
| Claim | Level | Reason | Evidence status | Confidence |
|---|---|---|---|---|
## 10. Incompleteness
[State what remains unproven, open, uncontrolled, or missing.]
## 11. Verdict
[State what the framework has earned, what it has not earned, and whether Extended analysis is warranted.]Flash is the fast mode.
Use it when you need a compact first read.
Apply the Flash Workbench to the provided theory, paper, or dispute.
Do not decide whether it is true. Classify its claims.
Return:
1. What kind of thing this is: fundamental candidate, reconstruction mechanism, consistency filter, effective theory, computational validator, observer framework, or endpoint phenomenology.
2. Its maturity stage: Seed, Model, Framework, Candidate Theory, or Serious Contender.
3. Main claim.
4. Claim type: mathematical, ontological, dynamical, observable, measurement, approximation, interpretation, or metaphor.
5. Explanation level: Derives, Explains, Accommodates, Postulates, or Undeclared.
6. What it forbids.
7. What would count against it.
8. Whether data can constrain its substrate or assumptions: Strong, Moderate, Weak, None, Undeclared, or Not Applicable.
9. Main incompleteness.
10. Flash verdict: what it has earned, what it has not earned, and whether Core or Extended analysis is needed.
Rules:
- Do not call accommodation derivation.
- Do not infer truth from author confidence.
- Do not treat metaphor as ontology.
- Mark missing fields explicitly.
- Use Not Applicable when the criterion does not fit the function class.
# Flash Workbench Report: [Target]
## 1. What it is
- Function class:
- Maturity stage:
## 2. Main claim
[One concise statement.]
## 3. Claim type
[Math / ontology / dynamics / observable / measurement / approximation / interpretation / metaphor]
## 4. Explanation level
Derives / Explains / Accommodates / Postulates / Undeclared
Reason:
## 5. What it forbids
- ...
## 6. What would count against it
- ...
## 7. Inverse constraint
Strong / Moderate / Weak / None / Undeclared / Not Applicable
Reason:
## 8. Main incompleteness
- ...
## 9. Flash verdict
- Earned:
- Not earned:
- Next step: No audit / Core audit / Extended auditExtended mode is for mature or high-stakes analysis.
Do not use Extended mode by default.
Use Extended when at least one condition holds:
- the framework claims fundamental unification;
- the framework claims to recover both spacetime/gravity and matter;
- the framework is being compared against major rivals;
- the analysis will be archived or reused;
- the claim is high-impact and likely to be overinterpreted;
- the Flash/Core result found serious category or falsifiability problems needing deeper audit.
Extended mode may add the following modules.
Use when the theory claims emergence or recovery of known physics.
reconstruction_map:
sector: geometry | gravity | matter | amplitudes | thermodynamics | observers | cosmology | other
domain:
codomain:
approximation_scheme:
control_parameter:
validity_regime:
failure_mode:
invariant_preserved:
inverse_constraint:
evidence_status:Use when equations, proofs, amplitudes, symmetries, or constraints are central.
formal_consistency:
well_formedness:
dimensional_consistency:
symmetry_or_anomaly_checks:
limit_recovery:
theorem_status: theorem | theorem_candidate | conjecture | heuristic | author_claim | unclear
unresolved_formal_gaps:Use when observations or experiments are central.
empirical_anchor:
tested_regime:
relevant_data:
agreement_claim:
tension_or_conflict:
possible_discriminating_signature:
accessibility: current | near_future | speculative | inaccessibleUse when comparing theories.
rival_adversarial_review:
rival:
strongest_objection_from_rival:
what_the_target_handles_better:
what_the_rival_handles_better:
shared_assumptions:
decisive_missing_test:Use S+ only when substrate-level inspection is needed.
S+ = (A, Ω, C, Φ, R, Σ, M, O)
| Slot | Meaning |
|---|---|
A |
Observable algebra / physically meaningful structures |
Ω |
Admissible states |
C |
Causal, relational, or ordering structure |
Φ |
Dynamics, amplitudes, constraints, or evolution |
R |
Coarse-graining, renormalization, emergence map |
Σ |
Symmetries, anomalies, consistency constraints |
M |
Reconstruction maps |
O |
Other component not captured above |
S+ is no longer the default interface. It is an Extended module.
The Minimal Viable Theory threshold applies only to frameworks that claim to be standalone fundamental or unification candidates.
Do not apply it to effective theories, reconstruction mechanisms, consistency filters, computational validators, observer frameworks, or endpoint phenomenology unless they claim fundamental status.
A serious fundamental candidate should address these seven criteria:
| # | Criterion | Question |
|---|---|---|
| 1 | Substrate and dynamics | What is fundamental, and how does it behave? |
| 2 | Known-limit recovery | Does it recover GR, QFT, thermodynamics, and relevant tested regimes? |
| 3 | Reconstruction control | Are emergence/recovery maps declared with domain, codomain, regime, and failure mode? |
| 4 | Consistency filters | Does it address anomalies, symmetries, EFT constraints, black-hole constraints, or relevant no-go results? |
| 5 | Inverse constraint | Can data or calculation constrain its substrate or assumptions? |
| 6 | Exclusion/prediction | What does it forbid or predict differently from rivals? |
| 7 | Incompleteness transparency | What can it not yet prove, and what would close the gap? |
Use simple statuses:
| Status | Meaning |
|---|---|
| Strong | Controlled, declared, and well supported. |
| Partial | Some coverage, but important gaps remain. |
| Weak | Mostly conjectural or undercontrolled. |
| Missing | Not declared. |
| Open | Explicitly unresolved with a reason. |
| Not Applicable | Wrong function class. |
The old diagnostic rules are replaced in the Core Workbench by compact risk flags.
Use these when triggered.
| Flag | Name | Trigger |
|---|---|---|
| R1 | Category Crossing | A claim moves from math/metaphor/interpretation to ontology without a bridge. |
| R2 | Explanation Upgrade | Accommodation or postulate is presented as explanation or derivation. |
| R3 | Missing Inverse Constraint | The theory cannot say how data constrains its substrate or assumptions. |
| R4 | Uncontrolled Emergence | A reconstruction claim lacks domain, codomain, regime, control parameter, or failure mode. |
| R5 | Empirical Drift | The theory lacks tested-regime recovery or empirical contact despite making physical claims. |
| R6 | Parameter Padding | A large choice space, model selection, or parameter tuning is used as if it were explanation. |
| R7 | Rival Mismatch | Frameworks of different function class are ranked as if they were direct competitors. |
| R8 | Overformalization | Early-stage speculation is being crushed by mature-theory requirements. |
| R9 | Source Laundering | Author rhetoric or LLM inference is treated as established result. |
| R10 | Completeness Pretense | Gaps, undecidable claims, or missing assumptions are hidden or minimized. |
Risk flags do not automatically refute a theory. They identify where caution is required.
Every important Workbench field should carry an evidence status.
Use this compact vocabulary:
| Evidence status | Meaning |
|---|---|
| explicit_source | The source directly states it. |
| equation_or_derivation | It follows from a displayed equation, derivation, or proof. |
| author_claim | The author claims it, but it is not independently established here. |
| reviewer_inference | The analyst infers it from the source. |
| community_context | It relies on background knowledge outside the source. |
| not_discussed | The source does not address it. |
| unsupported | The analyst or LLM cannot support it. |
Do not turn “the paper claims X” into “X is established.”
Keep these separate:
- The source says X.
- The analyst verifies that the source says X.
- The analyst judges whether X is established.
Every Core or Extended Workbench report must include incompleteness.
incompleteness:
unproven_core_claims:
-
open_questions:
-
missing_bridges:
-
uncontrolled_approximations:
-
missing_reconstruction_maps:
-
effect_on_main_claim:
what_would_close_the_gap:
-A theory that honestly declares its gaps is easier to evaluate than one that hides them.
The Workbench rewards honest incompleteness. It penalizes hidden incompleteness.
LLMs are clerks, not authorities.
- parse sources;
- extract claims;
- classify claim types;
- identify assumptions;
- draft Workbench reports;
- propose explanation levels;
- flag missing bridges;
- compare rival frameworks;
- generate adversarial objections.
- decide final truth;
- certify a theory;
- verify their own hallucinations;
- convert author claims into facts;
- replace physical intuition;
- replace expert review;
- silently fill missing fields.
The human reviewer must:
- verify source evidence;
- correct overclaims;
- adjudicate function class;
- approve explanation levels;
- judge inverse constraint strength;
- decide whether the analysis is fair.
The rule remains:
The human is the referee. The LLM is the clerk.
Before accepting a Workbench output, check:
| Check | Pass condition |
|---|---|
| Function class first | The output identifies what kind of framework is being analyzed. |
| Maturity stage declared | A Seed is not judged like a Serious Contender. |
| Claim type declared | Major claims are typed before being evaluated. |
| No truth verdict | The output does not declare the theory true or false. |
| No explanation upgrade | Accommodation is not upgraded to explanation or derivation. |
| Missing fields visible | Undeclared or not-discussed fields stay visible. |
| Bridge principles checked | Cross-type moves are not allowed without bridges. |
| Inverse constraint stated | The output says whether data can constrain substrate or assumptions. |
| Rival comparison fair | The theory is compared only against appropriate rivals. |
| Incompleteness included | Open gaps and closure conditions are stated. |
| Source status visible | Claims are not laundered from rhetoric into fact. |
| Adversarial pass done | The output has been re-read assuming overclaiming. |
Use the lightest workflow that fits the task.
- Identify function class.
- Identify maturity stage.
- State main claim.
- Type the claim.
- Classify explanation level.
- Ask what it forbids.
- Ask what would count against it.
- State incompleteness.
- Decide whether Core or Extended analysis is needed.
- Read source.
- Frame the physical problem.
- Identify function class and maturity stage.
- Extract claims.
- Type claims.
- List assumptions.
- Map constraints.
- Identify bridges and missing bridges.
- Evaluate inverse constraints.
- Compare against appropriate rivals.
- Declare incompleteness.
- Produce verdict.
- Run adversarial pass.
- Complete Core first.
- Add reconstruction maps where relevant.
- Add formal consistency checks.
- Add empirical anchor checks.
- Add rival adversarial modules.
- Apply MVT only if appropriate.
- Optionally apply S+ mapping.
- Archive report with evidence provenance.
Avoid or delay the Workbench when:
- the idea is pure brainstorming;
- the idea is too early even for claim typing;
- the goal is pedagogy rather than evaluation;
- the work is purely mathematical and makes no physical claim;
- the source material is too thin to support analysis;
- applying the method would prematurely kill a useful speculative seed;
- a lighter Flash pass is sufficient.
In early creative phases, use a gentler question:
What would this idea need to become Workbench-ready?
For everyday use, ask only these seven questions:
- What kind of thing is this?
- What is the main claim?
- What kind of claim is it?
- What assumptions does it require?
- What would break it?
- What does it explain better than rivals?
- What remains unproven?
This is the minimum Workbench.
Apply the Core Workbench to the provided theory, paper, or dispute.
Your task is not to decide whether it is true. Your task is to make its claims inspectable.
Return:
1. Metadata: source, domain, maturity stage, function class, confidence.
2. Central claim and scope.
3. Claim type: mathematical, ontology, dynamics, observable, measurement, approximation, interpretation, or metaphor.
4. Assumptions: explicit, hidden/inferred, inherited, speculative.
5. Constraint map: hard constraints, known physics to recover, consistency filters, empirical constraints, unaddressed constraints.
6. Bridge principles: what cross-type bridges are required and which are missing.
7. Inverse constraints: what it forbids, what would count against it, whether data can constrain substrate or assumptions.
8. Rival differential: what it explains better than appropriate rivals, what rivals explain better, and what remains unresolved.
9. Explanation ladder for major claims: Derives, Explains, Accommodates, Postulates, or Undeclared.
10. Incompleteness: unproven core claims, open questions, missing bridges, and what would close the gap.
11. Verdict: what it has earned, what it has not earned, strongest next test, and whether Extended analysis is warranted.
Rules:
- Do not judge truth.
- Do not call accommodation derivation.
- Do not treat metaphor as ontology.
- Do not compare unlike function classes as direct rivals.
- Use Not Applicable when a criterion does not fit.
- Mark missing fields explicitly.
- Separate author claims from established results.
- Run an adversarial pass before finalizing.
Review the previous Workbench output adversarially.
Assume the source, author, and LLM may have overclaimed.
Check:
1. Was function class identified before evaluation?
2. Was maturity stage respected?
3. Were claim types identified correctly?
4. Was any metaphor treated as ontology?
5. Was any accommodation upgraded to explanation or derivation?
6. Was any author claim treated as established result?
7. Were missing fields silently filled?
8. Were bridge principles required for cross-type moves?
9. Is the inverse constraint too generous?
10. Is the rival comparison apples-to-apples?
11. Are incompleteness and closure conditions explicit?
12. Are any risk flags triggered?
Revise conservatively. When uncertain, downgrade the explanation level or mark the field Missing, Open, Undeclared, or Not Applicable.
The compact Workbench clarifies that common “solutions” often occupy different function classes.
| Proposal | Function class | Compact clarification |
|---|---|---|
| Island formula | Reconstruction mechanism | Computes Page-curve behavior in a controlled semiclassical path-integral setting; does not by itself identify the microscopic substrate. |
| Fuzzballs | Fundamental candidate / microphysical proposal | Proposes black-hole microstate structure; strongest in controlled string-theory regimes; broader astrophysical generality remains open. |
| AMPS firewall | Consistency filter / diagnostic argument | Exposes tension among unitarity, EFT outside the horizon, and smooth horizons; not a positive ontology by itself. |
| Remnants | Endpoint phenomenology | Proposes a possible evaporation endpoint; must survive entropy, EFT, and observational consistency checks. |
The bad question is:
Which one solves the paradox?
The Workbench question is:
Which proposal supplies a controlled bridge from Page-curve behavior or black-hole observables back to microscopic structure, while surviving consistency filters and recovering semiclassical physics where required?
The compact Workbench is best understood as a constraint grammar for theoretical physics.
It does not discover new physics by itself. It does not certify truth. It does not replace expert judgment.
Its value is that it prevents avoidable confusion. It makes visible:
- what kind of framework is being discussed;
- what type of claim is being made;
- what assumptions are doing the work;
- what constraints must be survived;
- what bridges are missing;
- what the theory forbids;
- what would count against it;
- what it explains better than rivals;
- what remains unproven.
The compact form is:
Claim → Type → Assumptions → Constraints → Bridges → Inverse Constraints → Rival Differential
That is the essential Workbench.
Everything else is optional scaffolding.