eyeling 1.24.6 → 1.24.8

package/examples/output/{act-tunnel-junction-wake-switch.txt → act-tunnel-junction-wake-switch.md}

@@ -1,21 +1,23 @@
-ACT tunnel-junction wake switch
+# act-tunnel-junction-wake-switch  
 
-Answer
-YES for the tunnel junction.
-NO for the conventional low-bias PN junction in the same wake-switch regime.
+ACT tunnel-junction wake switch  
 
-Reason Why
-The tunnel junction is modeled as a heavily doped narrow PN junction with overlapping states, so quantum barrier transfer is possible. That makes sub-threshold current possible in the low-forward-bias regime, which in turn makes ultra-low-bias switching possible for the wake circuit. Because the device is also scanned through a peak-to-valley window, a negative differential response is possible as well. By contrast, the conventional junction lacks the structural conditions for the same transfer mode, so it cannot deliver the same low-bias switching task in this case.
+Answer  
+YES for the tunnel junction.  
+NO for the conventional low-bias PN junction in the same wake-switch regime.  
 
-Check
-C1  OK - the tunnel junction can support quantum barrier transfer
-C2  OK - the tunnel junction is classified as tunneling-dominant
-C3  OK - the tunnel junction can deliver sub-threshold current
-C4  OK - the tunnel junction can show negative differential response
-C5  OK - the tunnel junction can perform ultra-low-bias switching
-C6  OK - the tunnel junction can serve the leak-alarm wake circuit
-C7  OK - the conventional junction cannot support the same quantum barrier transfer
-C8  OK - the conventional junction cannot deliver sub-threshold current in this regime
-C9  OK - the conventional junction cannot show the tunnel-style negative differential response
-C10 OK - the conventional junction cannot perform ultra-low-bias switching here
-C11 OK - the conventional junction cannot serve the leak-alarm wake circuit in this case
+Reason Why  
+The tunnel junction is modeled as a heavily doped narrow PN junction with overlapping states, so quantum barrier transfer is possible. That makes sub-threshold current possible in the low-forward-bias regime, which in turn makes ultra-low-bias switching possible for the wake circuit. Because the device is also scanned through a peak-to-valley window, a negative differential response is possible as well. By contrast, the conventional junction lacks the structural conditions for the same transfer mode, so it cannot deliver the same low-bias switching task in this case.  
+
+Check  
+C1  OK - the tunnel junction can support quantum barrier transfer  
+C2  OK - the tunnel junction is classified as tunneling-dominant  
+C3  OK - the tunnel junction can deliver sub-threshold current  
+C4  OK - the tunnel junction can show negative differential response  
+C5  OK - the tunnel junction can perform ultra-low-bias switching  
+C6  OK - the tunnel junction can serve the leak-alarm wake circuit  
+C7  OK - the conventional junction cannot support the same quantum barrier transfer  
+C8  OK - the conventional junction cannot deliver sub-threshold current in this regime  
+C9  OK - the conventional junction cannot show the tunnel-style negative differential response  
+C10 OK - the conventional junction cannot perform ultra-low-bias switching here  
+C11 OK - the conventional junction cannot serve the leak-alarm wake circuit in this case  

package/examples/output/{act-yeast-self-reproduction.txt → act-yeast-self-reproduction.md}

@@ -1,23 +1,25 @@
-ACT yeast self-reproduction
+# act-yeast-self-reproduction  
 
-Answer
-YES for the viable starter culture.
-NO for accurate self-reproduction in the non-digital contrast lineage.
+ACT yeast self-reproduction  
 
-Reason Why
-The starter genome is treated as a replicator storing digital hereditary information, while the cell machinery is treated as the vehicle that enables metabolism and copying support. Under no-design laws, digital information makes accurate genome copying possible. Because the replicator is accurate and paired with a vehicle, the whole starter cell qualifies as a self-reproducer. With a variation source and a selection environment, natural selection also becomes possible. By contrast, the non-digital lineage cannot support accurate genome copying under the same no-design-laws assumption, so it cannot sustain the same accurate self-reproduction or natural-selection story.
+Answer  
+YES for the viable starter culture.  
+NO for accurate self-reproduction in the non-digital contrast lineage.  
 
-Check
-C1  OK - no-design laws are assumed
-C2  OK - digital information is physically instantiated for the viable lineage
-C3  OK - a viable replicator is present
-C4  OK - a viable vehicle is present
-C5  OK - accurate genome copying is possible for the viable lineage
-C6  OK - the viable lineage has a replicator-plus-vehicle architecture
-C7  OK - the starter cell qualifies as a self-reproducer
-C8  OK - heritable variation is possible for the viable lineage
-C9  OK - the viable starter culture can support a natural-selection lineage
-C10 OK - the fitter viable variant is selected in the fermenter
-C11 OK - the non-digital contrast genome cannot be copied accurately under no-design laws
-C12 OK - the non-digital contrast cell cannot achieve accurate self-reproduction
-C13 OK - the non-digital contrast cell cannot support the same natural-selection lineage
+Reason Why  
+The starter genome is treated as a replicator storing digital hereditary information, while the cell machinery is treated as the vehicle that enables metabolism and copying support. Under no-design laws, digital information makes accurate genome copying possible. Because the replicator is accurate and paired with a vehicle, the whole starter cell qualifies as a self-reproducer. With a variation source and a selection environment, natural selection also becomes possible. By contrast, the non-digital lineage cannot support accurate genome copying under the same no-design-laws assumption, so it cannot sustain the same accurate self-reproduction or natural-selection story.  
+
+Check  
+C1  OK - no-design laws are assumed  
+C2  OK - digital information is physically instantiated for the viable lineage  
+C3  OK - a viable replicator is present  
+C4  OK - a viable vehicle is present  
+C5  OK - accurate genome copying is possible for the viable lineage  
+C6  OK - the viable lineage has a replicator-plus-vehicle architecture  
+C7  OK - the starter cell qualifies as a self-reproducer  
+C8  OK - heritable variation is possible for the viable lineage  
+C9  OK - the viable starter culture can support a natural-selection lineage  
+C10 OK - the fitter viable variant is selected in the fermenter  
+C11 OK - the non-digital contrast genome cannot be copied accurately under no-design laws  
+C12 OK - the non-digital contrast cell cannot achieve accurate self-reproduction  
+C13 OK - the non-digital contrast cell cannot support the same natural-selection lineage  

package/examples/output/annotation.md

@@ -0,0 +1 @@
+# annotation  

package/examples/output/auroracare.md

@@ -0,0 +1,150 @@
+# auroracare  
+
+AuroraCare — Purpose-based Medical Data Exchange  
+
+## A – Primary care visit  
+Clinician in the patient's care team accessing the patient summary for primary care management.  
+
+Answer  
+PERMIT  
+
+Reason Why  
+Permitted: clinician in the patient's care team, and the primary-care policy matched.  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 OK - clinician  
+C3 OK - care-team linked  
+C4 SKIPPED  
+C5 OK - operator=isAnyOf, allowed=["https://example.org/health#PATIENT_SUMMARY","https://example.org/health#LAB_RESULTS"], requested=["https://example.org/health#PATIENT_SUMMARY"]  
+C6 SKIPPED - no prohibition matched  
+C7 OK - trace shows matching permission  
+C8 SKIPPED - no matched policy or no duties  
+C9 SKIPPED - policy has no environment constraint  
+C10 INFO - matched policy: urn:policy:primary-care-001  
+
+## B – Quality improvement (in scope)  
+QI analyst using lab results + summary in a secure environment.  
+
+Answer  
+PERMIT  
+
+Reason Why  
+Permitted: ODRL/DPV policy matched for secondary use.  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 SKIPPED  
+C3 SKIPPED  
+C4 OK - opt-in present and policy matched  
+C5 OK - operator=isAllOf, allowed=["https://example.org/health#LAB_RESULTS","https://example.org/health#PATIENT_SUMMARY"], requested=["https://example.org/health#LAB_RESULTS","https://example.org/health#PATIENT_SUMMARY"]  
+C6 SKIPPED - no prohibition matched  
+C7 OK - trace shows matching permission  
+C8 INFO - duties attached: duty:https://w3id.org/dpv/legal/eu/ehds#requireConsent, duty:https://w3id.org/dpv/legal/eu/ehds#noExfiltration  
+C9 OK - operator=eq, allowed="secure_env", requested="secure_env"  
+C10 INFO - matched policy: urn:policy:qi-2025-aurora  
+
+## C – Quality improvement (out of scope)  
+QI analyst with only lab results; policy expects labs + summary.  
+
+Answer  
+DENY  
+
+Reason Why  
+Denied: no policy matched (purpose, environment, TOMs, or categories out of scope).  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 SKIPPED  
+C3 SKIPPED  
+C4 OK - denied because opt-in missing or no policy match  
+C5 SKIPPED  
+C6 SKIPPED - no prohibition matched  
+C7 SKIPPED  
+C8 SKIPPED - no matched policy or no duties  
+C9 SKIPPED  
+C10 SKIPPED - no matched policy  
+
+## D – Insurance management  
+Insurance bot attempting to use health data for insurance management (prohibited purpose).  
+
+Answer  
+DENY  
+
+Reason Why  
+Denied: the requested purpose (insurance management) is prohibited by policy.  
+
+Check  
+C1 OK - denied prohibited purpose  
+C2 SKIPPED  
+C3 SKIPPED  
+C4 SKIPPED  
+C5 SKIPPED  
+C6 OK - denied due to prohibition  
+C7 SKIPPED  
+C8 SKIPPED - no matched policy or no duties  
+C9 SKIPPED  
+C10 SKIPPED - no matched policy  
+
+## E – GP checks labs  
+GP for the same patient checking lab results via the API gateway.  
+
+Answer  
+PERMIT  
+
+Reason Why  
+Permitted: clinician in the patient's care team, and the primary-care policy matched.  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 OK - clinician  
+C3 OK - care-team linked  
+C4 SKIPPED  
+C5 OK - operator=isAnyOf, allowed=["https://example.org/health#PATIENT_SUMMARY","https://example.org/health#LAB_RESULTS"], requested=["https://example.org/health#LAB_RESULTS"]  
+C6 SKIPPED - no prohibition matched  
+C7 OK - trace shows matching permission  
+C8 SKIPPED - no matched policy or no duties  
+C9 SKIPPED - policy has no environment constraint  
+C10 INFO - matched policy: urn:policy:primary-care-001  
+
+## F – Research on anonymised dataset  
+Researcher using anonymised labs + summary in a secure environment, with opt-in.  
+
+Answer  
+PERMIT  
+
+Reason Why  
+Permitted: subject opted in and an ODRL/DPV policy matched (anonymised dataset in secure environment).  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 SKIPPED  
+C3 SKIPPED  
+C4 OK - opt-in present and policy matched  
+C5 OK - operator=isAnyOf, allowed=["https://example.org/health#LAB_RESULTS","https://example.org/health#PATIENT_SUMMARY","https://example.org/health#IMAGING_REPORT"], requested=["https://example.org/health#PATIENT_SUMMARY","https://example.org/health#LAB_RESULTS"]  
+C6 SKIPPED - no prohibition matched  
+C7 OK - trace shows matching permission  
+C8 INFO - duties attached: duty:https://w3id.org/dpv/legal/eu/ehds#annualOutcomeReport, duty:https://w3id.org/dpv/legal/eu/ehds#noReidentification, duty:https://w3id.org/dpv/legal/eu/ehds#noExfiltration  
+C9 OK - operator=eq, allowed="secure_env", requested="secure_env"  
+C10 INFO - matched policy: urn:policy:research-aurora-diabetes  
+
+## G – AI training (opt-out)  
+Data user wants to train AI, but the subject opted out of AI training.  
+
+Answer  
+DENY  
+
+Reason Why  
+Denied: you opted out of your data being used to train AI systems.  
+
+Check  
+C1 SKIPPED - not a prohibited purpose  
+C2 SKIPPED  
+C3 SKIPPED  
+C4 OK - denied because opt-in missing or no policy match  
+C5 SKIPPED  
+C6 SKIPPED - no prohibition matched  
+C7 SKIPPED  
+C8 SKIPPED - no matched policy or no duties  
+C9 SKIPPED  
+C10 SKIPPED - no matched policy  

package/examples/output/backward-recursion.md

@@ -0,0 +1,6 @@
+# backward-recursion  
+
+@prefix : <urn:example#> .  
+
+:a :reaches :b .  
+:a :reaches :c .  

package/examples/output/barley-seed-becoming.md

@@ -0,0 +1,27 @@
+# barley-seed-becoming  
+
+Barley seed lineage — becoming  
+
+Answer  
+YES for the viable barley lineage.  
+NO for the contrast lineages when digital heredity, repair, protected dormancy, or heritable variation are missing.  
+
+Reason Why  
+The main lineage can be read as a becoming: a protected dormant seed can germinate, an adult plant can become a next seed stage, and the lineage can therefore become a self-renewing cycle. Because its hereditary information is digitally instantiated and repair is available, it can also become an accurately reproduced next generation under no-design laws. And because heritable variation is present under a matching selection environment, it can become an adaptively persistent lineage. The contrast lineages mark blocked becomings: non-digital heredity blocks accurate copying, lack of repair blocks reliable renewal, lack of dormancy protection blocks closure through the seed phase, and lack of heritable variation blocks adaptive becoming.  
+
+Check  
+B1  OK - no-design laws are assumed  
+B2  OK - the viable genome can become accurately copied  
+B3  OK - the viable seed can become a protected dormant phase  
+B4  OK - the viable seed can become a germinating stage  
+B5  OK - the viable adult can become a next dormant seed stage  
+B6  OK - the viable lineage can become an accurately reproduced next generation  
+B7  OK - the viable lineage can become a closed life cycle  
+B8  OK - the viable lineage can become a novel variant lineage  
+B9  OK - the viable lineage can become adaptively persistent  
+B10 OK - the viable lineage is an evolvable becoming  
+B11 OK - the non-digital lineage cannot become an accurately reproduced next generation  
+B12 OK - the repair-deficient lineage cannot become an accurately reproduced next generation  
+B13 OK - the coatless lineage cannot become a closed life cycle  
+B14 OK - the static lineage cannot become an adaptive lineage  
+B15 OK - the static lineage cannot become an evolvable becoming  

package/examples/output/{bmi.txt → bmi.md}

@@ -1,20 +1,22 @@
-BMI — ARC-style Body Mass Index example
+# bmi  
 
-Answer
-BMI = 22.72
-Category = Normal
-At height 178 cm, a healthy-weight range is about 58.6–78.9 kg (BMI 18.5–24.9).
+BMI — ARC-style Body Mass Index example  
 
-Reason Why
-BMI is defined as weight in kilograms divided by height in meters squared. This program first normalizes the input to SI units, computes BMI, and then applies WHO adult categories as half-open intervals. The healthy-weight band is the weight range at the same height that corresponds to BMI 18.5 through 24.9.
+Answer  
+BMI = 22.72  
+Category = Normal  
+At height 178 cm, a healthy-weight range is about 58.6–78.9 kg (BMI 18.5–24.9).  
 
-Check
-C1 OK - the input was normalized into positive SI values.
-C2 OK - height squared was reconstructed from the normalized height.
-C3 OK - the BMI value matches the BMI = kg / m² formula.
-C4 OK - a BMI of 18.49 stays below the normal-weight threshold.
-C5 OK - the lower boundary is half-open: BMI 18.5 is classified as Normal.
-C6 OK - BMI 25.0 starts the Overweight category.
-C7 OK - BMI 30.0 starts the Obesity I category.
-C8 OK - classification behavior is monotonic across representative BMI values.
-C9 OK - the healthy-weight band was reconstructed from BMI 18.5 to 24.9 at the same height.
+Reason Why  
+BMI is defined as weight in kilograms divided by height in meters squared. This program first normalizes the input to SI units, computes BMI, and then applies WHO adult categories as half-open intervals. The healthy-weight band is the weight range at the same height that corresponds to BMI 18.5 through 24.9.  
+
+Check  
+C1 OK - the input was normalized into positive SI values.  
+C2 OK - height squared was reconstructed from the normalized height.  
+C3 OK - the BMI value matches the BMI = kg / m² formula.  
+C4 OK - a BMI of 18.49 stays below the normal-weight threshold.  
+C5 OK - the lower boundary is half-open: BMI 18.5 is classified as Normal.  
+C6 OK - BMI 25.0 starts the Overweight category.  
+C7 OK - BMI 30.0 starts the Obesity I category.  
+C8 OK - classification behavior is monotonic across representative BMI values.  
+C9 OK - the healthy-weight band was reconstructed from BMI 18.5 to 24.9 at the same height.  

package/examples/output/builtin-coverage.md

@@ -0,0 +1 @@
+# builtin-coverage  

package/examples/output/calidor.md

@@ -0,0 +1,31 @@
+# calidor  
+
+## Answer  
+The city is allowed to use a narrow heatwave-response insight and recommends Calidor Priority Cooling Bundle for this household.  
+case                 : calidor  
+decision             : Allowed  
+municipality         : Calidor  
+recommended package  : Calidor Priority Cooling Bundle  
+
+## Reason Why  
+The gateway desensitizes local heat, vulnerability, and prepaid-energy stress into an expiring municipal support insight, and the city consumes that envelope only for heatwave response.  
+metric               : active_need_count  
+threshold            : 3.0  
+scope                : household-gateway @ heat-alert-window  
+required capabilities: bill_credit, cooling_kit, transport, welfare_check  
+signature alg        : HMAC-SHA256  
+expires at           : 2026-07-18T21:00:00+00:00  
+reason.txt           : The gateway keeps raw indoor heat, vulnerability, and prepaid-energy data local, derives a priority-support signal, and shares only a scoped heatwave-response envelope with expiry.  
+dispatches logged    : 1  
+
+## Check  
+signature verifies              : yes  
+payload hash matches            : yes  
+minimization strips sensitive terms: yes  
+scope complete                  : yes  
+authorization allowed           : yes  
+heat-alert active               : yes  
+unsafe indoor heat              : yes  
+recommended package eligible    : yes  
+duty timing consistent          : yes  
+tenant screening prohibited     : yes  

package/examples/output/collection.md

@@ -0,0 +1 @@
+# collection  

package/examples/output/{complex-matrix-stability.txt → complex-matrix-stability.md}

@@ -1,14 +1,16 @@
-Complex Matrix Stability — ARC-style
+# complex-matrix-stability  
 
-Answer
-We compare three diagonal 2x2 complex matrices for discrete-time stability: A_unstable = [[(1,1),(0,0)],[(0,0),(2,0)]], A_stable = [[(1,0),(0,0)],[(0,0),(-1,0)]], and A_damped = [[(0,0),(0,0)],[(0,0),(0,0)]]. Their spectral radii are ρ(A_unstable) = 2, ρ(A_stable) = 1, and ρ(A_damped) = 0. So A_unstable is unstable, A_stable is marginally stable, and A_damped is damped.
+Complex Matrix Stability — ARC-style  
 
-Reason Why
-For a discrete-time linear system x_{k+1} = A x_k, the eigenvalues of A govern the behaviour of the modes. Because these matrices are diagonal, the eigenvalues are just the diagonal entries. The spectral radius is the maximum modulus of the eigenvalues: if it is greater than 1 a mode grows, if it equals 1 the modes remain bounded without decaying, and if it is less than 1 all modes decay to zero. Here the diagonal entries give radii 2, 1, and 0 respectively, which explains the three classifications.
+Answer  
+We compare three diagonal 2x2 complex matrices for discrete-time stability: A_unstable = [[(1,1),(0,0)],[(0,0),(2,0)]], A_stable = [[(1,0),(0,0)],[(0,0),(-1,0)]], and A_damped = [[(0,0),(0,0)],[(0,0),(0,0)]]. Their spectral radii are ρ(A_unstable) = 2, ρ(A_stable) = 1, and ρ(A_damped) = 0. So A_unstable is unstable, A_stable is marginally stable, and A_damped is damped.  
 
-Check
-C1 OK - A_unstable has eigenvalues (1,1) and (2,0) with spectral radius 2, so it is unstable.
-C2 OK - A_stable has eigenvalues (1,0) and (-1,0) with spectral radius 1, so it is marginally stable.
-C3 OK - A_damped has eigenvalues (0,0) and (0,0) with spectral radius 0, so every mode decays to zero.
-C4 OK - for z = (1,2) and w = (0,1), the squared modulus of z*w equals the product of the squared moduli.
-C5 OK - the spectral-radius-squared of 2*A_unstable is four times that of A_unstable.
+Reason Why  
+For a discrete-time linear system x_{k+1} = A x_k, the eigenvalues of A govern the behaviour of the modes. Because these matrices are diagonal, the eigenvalues are just the diagonal entries. The spectral radius is the maximum modulus of the eigenvalues: if it is greater than 1 a mode grows, if it equals 1 the modes remain bounded without decaying, and if it is less than 1 all modes decay to zero. Here the diagonal entries give radii 2, 1, and 0 respectively, which explains the three classifications.  
+
+Check  
+C1 OK - A_unstable has eigenvalues (1,1) and (2,0) with spectral radius 2, so it is unstable.  
+C2 OK - A_stable has eigenvalues (1,0) and (-1,0) with spectral radius 1, so it is marginally stable.  
+C3 OK - A_damped has eigenvalues (0,0) and (0,0) with spectral radius 0, so every mode decays to zero.  
+C4 OK - for z = (1,2) and w = (0,1), the squared modulus of z*w equals the product of the squared moduli.  
+C5 OK - the spectral-radius-squared of 2*A_unstable is four times that of A_unstable.  

package/examples/output/context-association.md

@@ -0,0 +1,7 @@
+# Context association  
+
+## Entailment  
+The RDF dataset associates Bob's data graph with a Data Integrity proof graph and a second metadata proof graph.  
+
+## Explanation  
+The input TriG names three graph contexts. The data graph states Bob's name. The signature graph links to that data graph with a proof and records an ecdsa-rdfc-2019 Data Integrity proof from the university issuer. The metadata graph then signs the signature graph itself, giving a chained context association.  

package/examples/output/{control-system.txt → control-system.md}

@@ -1,20 +1,22 @@
-Control System — ARC explanation of two control signals
+# control-system  
 
-Answer
-Send both actuator commands now.
-Actuator 1 command: 39.27346198678276
-Actuator 2 command: 26.08
+Control System — ARC explanation of two control signals  
 
-Reason Why
-The first sensor pair is 6 and 11, so the reading is rising and the controller normalizes the gap 5 into 2.23606797749979. That normalized value creates a feedforward term of 43.82693235899588, while the known disturbance 35766 contributes a compensation term of 4.553470372213121. Subtracting that compensation gives actuator 1 the command 39.27346198678276. For actuator 2, the target is 5 units above the measured output, so the tracking error is positive. The observed state is -2 relative units below the measured output, so the differential correction is negative. That yields a proportional feedback part of 29, a nonlinear factor of 1.46, and a differential contribution of -2.92. Together they produce actuator 2 command 26.08.
+Answer  
+Send both actuator commands now.  
+Actuator 1 command: 39.27346198678276  
+Actuator 2 command: 26.08  
 
-Check
-C1 OK - the first sensor pair is rising, so the normalization uses the rising-branch rule.
-C2 OK - the normalized measurement is positive and smaller than the raw gap, which is consistent with a square-root normalization.
-C3 OK - actuator 1 is lower than its proportional feedforward term because disturbance compensation is subtracted.
-C4 OK - the target is above the measured output, so the tracking error is positive.
-C5 OK - the observed state is below the measured output, so the differential error is negative.
-C6 OK - actuator 2 is lower than its pure proportional term because the differential part reduces it.
-C7 OK - actuator 1 matches an independently reconstructed feedforward calculation.
-C8 OK - actuator 2 matches an independently reconstructed feedback calculation.
-C9 OK - both actuator commands stay positive.
+Reason Why  
+The first sensor pair is 6 and 11, so the reading is rising and the controller normalizes the gap 5 into 2.23606797749979. That normalized value creates a feedforward term of 43.82693235899588, while the known disturbance 35766 contributes a compensation term of 4.553470372213121. Subtracting that compensation gives actuator 1 the command 39.27346198678276. For actuator 2, the target is 5 units above the measured output, so the tracking error is positive. The observed state is -2 relative units below the measured output, so the differential correction is negative. That yields a proportional feedback part of 29, a nonlinear factor of 1.46, and a differential contribution of -2.92. Together they produce actuator 2 command 26.08.  
+
+Check  
+C1 OK - the first sensor pair is rising, so the normalization uses the rising-branch rule.  
+C2 OK - the normalized measurement is positive and smaller than the raw gap, which is consistent with a square-root normalization.  
+C3 OK - actuator 1 is lower than its proportional feedforward term because disturbance compensation is subtracted.  
+C4 OK - the target is above the measured output, so the tracking error is positive.  
+C5 OK - the observed state is below the measured output, so the differential error is negative.  
+C6 OK - actuator 2 is lower than its pure proportional term because the differential part reduces it.  
+C7 OK - actuator 1 matches an independently reconstructed feedforward calculation.  
+C8 OK - actuator 2 matches an independently reconstructed feedback calculation.  
+C9 OK - both actuator commands stay positive.  

package/examples/output/{deep-taxonomy-10.txt → deep-taxonomy-10.md}

@@ -1,15 +1,17 @@
-Deep Taxonomy - deep classification benchmark
+# deep-taxonomy-10  
 
-Answer
-The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N10 and then as A2.
+Deep Taxonomy - deep classification benchmark  
 
-Reason Why
-Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N10, the final rule deriving A2 fires, and that in turn makes the test true.
+Answer  
+The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N10 and then as A2.  
 
-Check
-C1 OK - the starting classification N0 is present.
-C2 OK - the first expansion produced N1 together with side labels I1 and J1.
-C3 OK - the chain reaches the midpoint N5 and still carries both side-label branches.
-C4 OK - the final taxonomy step from N9 to N10 was completed.
-C5 OK - once N10 is reached, the terminal class A2 is derived.
-C6 OK - the success flag is raised only after the terminal class A2 is present.
+Reason Why  
+Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N10, the final rule deriving A2 fires, and that in turn makes the test true.  
+
+Check  
+C1 OK - the starting classification N0 is present.  
+C2 OK - the first expansion produced N1 together with side labels I1 and J1.  
+C3 OK - the chain reaches the midpoint N5 and still carries both side-label branches.  
+C4 OK - the final taxonomy step from N9 to N10 was completed.  
+C5 OK - once N10 is reached, the terminal class A2 is derived.  
+C6 OK - the success flag is raised only after the terminal class A2 is present.  

package/examples/output/{deep-taxonomy-100.txt → deep-taxonomy-100.md}

@@ -1,15 +1,17 @@
-Deep Taxonomy - deep classification benchmark
+# deep-taxonomy-100  
 
-Answer
-The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N100 and then as A2.
+Deep Taxonomy - deep classification benchmark  
 
-Reason Why
-Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N100, the final rule deriving A2 fires, and that in turn makes the test true.
+Answer  
+The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N100 and then as A2.  
 
-Check
-C1 OK - the starting classification N0 is present.
-C2 OK - the first expansion produced N1 together with side labels I1 and J1.
-C3 OK - the chain reaches the midpoint N50 and still carries both side-label branches.
-C4 OK - the final taxonomy step from N99 to N100 was completed.
-C5 OK - once N100 is reached, the terminal class A2 is derived.
-C6 OK - the success flag is raised only after the terminal class A2 is present.
+Reason Why  
+Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N100, the final rule deriving A2 fires, and that in turn makes the test true.  
+
+Check  
+C1 OK - the starting classification N0 is present.  
+C2 OK - the first expansion produced N1 together with side labels I1 and J1.  
+C3 OK - the chain reaches the midpoint N50 and still carries both side-label branches.  
+C4 OK - the final taxonomy step from N99 to N100 was completed.  
+C5 OK - once N100 is reached, the terminal class A2 is derived.  
+C6 OK - the success flag is raised only after the terminal class A2 is present.  

package/examples/output/{deep-taxonomy-1000.txt → deep-taxonomy-1000.md}

@@ -1,15 +1,17 @@
-Deep Taxonomy - deep classification benchmark
+# deep-taxonomy-1000  
 
-Answer
-The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N1000 and then as A2.
+Deep Taxonomy - deep classification benchmark  
 
-Reason Why
-Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N1000, the final rule deriving A2 fires, and that in turn makes the test true.
+Answer  
+The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N1000 and then as A2.  
 
-Check
-C1 OK - the starting classification N0 is present.
-C2 OK - the first expansion produced N1 together with side labels I1 and J1.
-C3 OK - the chain reaches the midpoint N500 and still carries both side-label branches.
-C4 OK - the final taxonomy step from N999 to N1000 was completed.
-C5 OK - once N1000 is reached, the terminal class A2 is derived.
-C6 OK - the success flag is raised only after the terminal class A2 is present.
+Reason Why  
+Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N1000, the final rule deriving A2 fires, and that in turn makes the test true.  
+
+Check  
+C1 OK - the starting classification N0 is present.  
+C2 OK - the first expansion produced N1 together with side labels I1 and J1.  
+C3 OK - the chain reaches the midpoint N500 and still carries both side-label branches.  
+C4 OK - the final taxonomy step from N999 to N1000 was completed.  
+C5 OK - once N1000 is reached, the terminal class A2 is derived.  
+C6 OK - the success flag is raised only after the terminal class A2 is present.  

package/examples/output/{deep-taxonomy-10000.txt → deep-taxonomy-10000.md}

@@ -1,15 +1,17 @@
-Deep Taxonomy - deep classification benchmark
+# deep-taxonomy-10000  
 
-Answer
-The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N10000 and then as A2.
+Deep Taxonomy - deep classification benchmark  
 
-Reason Why
-Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N10000, the final rule deriving A2 fires, and that in turn makes the test true.
+Answer  
+The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N10000 and then as A2.  
 
-Check
-C1 OK - the starting classification N0 is present.
-C2 OK - the first expansion produced N1 together with side labels I1 and J1.
-C3 OK - the chain reaches the midpoint N5000 and still carries both side-label branches.
-C4 OK - the final taxonomy step from N9999 to N10000 was completed.
-C5 OK - once N10000 is reached, the terminal class A2 is derived.
-C6 OK - the success flag is raised only after the terminal class A2 is present.
+Reason Why  
+Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N10000, the final rule deriving A2 fires, and that in turn makes the test true.  
+
+Check  
+C1 OK - the starting classification N0 is present.  
+C2 OK - the first expansion produced N1 together with side labels I1 and J1.  
+C3 OK - the chain reaches the midpoint N5000 and still carries both side-label branches.  
+C4 OK - the final taxonomy step from N9999 to N10000 was completed.  
+C5 OK - once N10000 is reached, the terminal class A2 is derived.  
+C6 OK - the success flag is raised only after the terminal class A2 is present.  

package/examples/output/{deep-taxonomy-100000.txt → deep-taxonomy-100000.md}

@@ -1,15 +1,17 @@
-Deep Taxonomy - very deep classification benchmark
+# deep-taxonomy-100000  
 
-Answer
-The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N100000 and then as A2.
+Deep Taxonomy - very deep classification benchmark  
 
-Reason Why
-Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N100000, the final rule deriving A2 fires, and that in turn makes the test true.
+Answer  
+The test succeeds: starting from one individual classified as N0, the rules eventually classify it as N100000 and then as A2.  
 
-Check
-C1 OK - the starting classification N0 is present.
-C2 OK - the first expansion produced N1 together with side labels I1 and J1.
-C3 OK - the chain reaches the midpoint N50000 and still carries both side-label branches.
-C4 OK - the final taxonomy step from N99999 to N100000 was completed.
-C5 OK - once N100000 is reached, the terminal class A2 is derived.
-C6 OK - the success flag is raised only after the terminal class A2 is present.
+Reason Why  
+Each rule moves the same individual one level deeper in the taxonomy and also adds two side labels. Because that chain continues all the way from N0 to N100000, the final rule deriving A2 fires, and that in turn makes the test true.  
+
+Check  
+C1 OK - the starting classification N0 is present.  
+C2 OK - the first expansion produced N1 together with side labels I1 and J1.  
+C3 OK - the chain reaches the midpoint N50000 and still carries both side-label branches.  
+C4 OK - the final taxonomy step from N99999 to N100000 was completed.  
+C5 OK - once N100000 is reached, the terminal class A2 is derived.  
+C6 OK - the success flag is raised only after the terminal class A2 is present.