eyeling 1.33.6 → 1.33.8

package/README.md

@@ -828,8 +828,8 @@ The eyelang engine has its own built-in registry under `lib/eyelang/builtins/`.
 | Lists | `append/3`, `nth0/3`, `set_nth0/4`, `rest/2`, `member/2`, `select/3`, `not_member/2`, `reverse/2`, `length/2`, `sort/2` |
 | Aggregation | `findall/3`, `countall/2`, `sumall/3`, `aggregate_min/5`, `aggregate_max/5` |
 | Control | `not/1`, `once/1` |
-| Formula terms | `formula_binary/4` |
-| Search and optimization helpers | `n_queens/2`, `weighted_hamiltonian_cycle/4`, `weighted_hamiltonian_path/4`, `hamiltonian_cycle/3`, `fixed_length_cycle/4`, `bounded_path/5`, `bounded_subset/7`, `cnf_model/3`, `qm_prime_implicants/4`, `qm_minimal_cover/4`, `alphametic_sum/5` |
+| Context terms | `holds/2`, `holds/3` |
+| Search and optimization helpers | `n_queens/2`, `weighted_hamiltonian_cycle/4`, `weighted_hamiltonian_path/4`, `hamiltonian_cycle/3`, `fixed_length_cycle/4`, `bounded_path/5`, `cnf_model/3`, `qm_prime_implicants/4`, `qm_minimal_cover/4`, `alphametic_sum/5` |
 | Numeric extension helpers | `extended_gcd/5`, `collatz_trajectory/2`, `kaprekar_steps/2`, `goldbach_pair/3` |
 | Matrix helpers | `matrix_sum/2`, `matrix_multiply/2`, `cholesky_decomposition/2`, `determinant/2`, `matrix_inv_triang/2`, `matrix_inversion/2` |
 

package/docs/eyelang-guide.md

@@ -16,7 +16,7 @@ For the normative language definition, including lexical syntax, terms, clauses,
 3. [Default output](#default-output)
 4. [Writing programs](#writing-programs)
 5. [Aggregation helpers](#aggregation-helpers)
-6. [Formula data](#formula-data)
+6. [Context data](#context-data)
 7. [Example catalog](#example-catalog)
 8. [Golden outputs, tests, and conformance](#golden-outputs-tests-and-conformance)
 9. [Development and release](#development-and-release)
@@ -250,7 +250,7 @@ The playground has matching `--stats` and `--proof` checkboxes, so browser runs
 
 eyelang builtins are registered by name and arity in small modules under [`lib/eyelang/builtins`](../lib/eyelang/builtins). This keeps the runtime portable to Node.js and the browser while giving each builtin family a clear boundary. Builtins are enabled by normal predicate calls.
 
-The core builtin families cover unification, arithmetic, comparison, dates, strings, lists, aggregation, formula terms, and search control. Additional reusable finite-search helpers are available only where bundled examples need them to avoid large amounts of repetitive generate-and-test code. These helpers are deliberately general relations rather than shortcuts tied to a particular example name. For example:
+The core builtin families cover unification, arithmetic, comparison, dates, strings, lists, aggregation, context terms, and search control. Additional reusable finite-search helpers are available only where bundled examples need them to avoid large amounts of repetitive generate-and-test code. These helpers are deliberately general relations rather than shortcuts tied to a particular example name. For example:
 
 ```prolog
 answer(Queens) :-
@@ -287,17 +287,16 @@ best_cycle(Cycle, Cost) :-
   aggregate_min([Cost, Cycle], Cycle, candidate_cycle(Cities, Cycle, Cost), [Cost, Cycle], Cycle).
 ```
 
-## Formula data
+## Context data
 
-Comma terms can be data as well as conjunctions. eyelang provides a relation-oriented formula utility.
-
-`formula_binary(Formula, S, P, O)` enumerates binary terms and exposes their functor as an atom constant:
+Comma terms can be data as well as conjunctions. eyelang provides two context utilities:
 
 ```prolog
-formula_binary((name(alice, "Alice"), knows(alice, bob)), S, P, O).
+holds((name(alice, "Alice"), knows(alice, bob)), name(S, O)).
+holds((ready, name(alice, "Alice"), route(alice, bob, 7)), Name, Args).
 ```
 
-This can yield `S = alice`, `P = name`, `O = "Alice"` and `S = alice`, `P = knows`, `O = bob`. The utility is useful for quoted formula data, but it does not make those formula members true in the ambient program.
+Use `holds/2` when you want to match the member term directly, for example `name(S, O)`, `route(A, B, Cost)`, or `edge(A, arc(B, Cost))`. Use `holds/3` when you need the predicate name and argument list as data: it exposes any-arity member as atom constant `Name` plus a proper list `Args`, so zero-, binary-, and ternary members appear as `ready/0`, `name/2`, and `route/3` shapes without a special binary predicate. These utilities are useful for quoted context data, but they do not make those context members true in the ambient program.
 
 
 ## Example catalog
@@ -416,12 +415,10 @@ The repository includes examples for recursion, graph reachability, finite searc
 | [`peasant.pl`](../examples/eyelang/peasant.pl) | Performs peasant multiplication and exponentiation. | [`output/peasant.pl`](../examples/eyelang/output/peasant.pl) |
 | [`pendulum-period.pl`](../examples/eyelang/pendulum-period.pl) | Computes simple pendulum periods. | [`output/pendulum-period.pl`](../examples/eyelang/output/pendulum-period.pl) |
 | [`polynomial.pl`](../examples/eyelang/polynomial.pl) | Finds complex integer polynomial roots. | [`output/polynomial.pl`](../examples/eyelang/output/polynomial.pl) |
-| [`project-portfolio-optimization.pl`](../examples/eyelang/project-portfolio-optimization.pl) | Optimizes a constrained project portfolio with pruning and aggregate builtins. | [`output/project-portfolio-optimization.pl`](../examples/eyelang/output/project-portfolio-optimization.pl) |
 | [`proof-contrapositive.pl`](../examples/eyelang/proof-contrapositive.pl) | Models proof by contrapositive. | [`output/proof-contrapositive.pl`](../examples/eyelang/output/proof-contrapositive.pl) |
 | [`quadratic-formula.pl`](../examples/eyelang/quadratic-formula.pl) | Solves sample quadratic equations. | [`output/quadratic-formula.pl`](../examples/eyelang/output/quadratic-formula.pl) |
 | [`quine-mccluskey.pl`](../examples/eyelang/quine-mccluskey.pl) | Minimizes Boolean terms with Quine-McCluskey. | [`output/quine-mccluskey.pl`](../examples/eyelang/output/quine-mccluskey.pl) |
 | [`radioactive-decay.pl`](../examples/eyelang/radioactive-decay.pl) | Computes radioactive decay over time. | [`output/radioactive-decay.pl`](../examples/eyelang/output/radioactive-decay.pl) |
-| [`resilient-city-orchestration.pl`](../examples/eyelang/resilient-city-orchestration.pl) | Orchestrates storm-response missions from signals, policy, routes, teams, and portfolio optimization. | [`output/resilient-city-orchestration.pl`](../examples/eyelang/output/resilient-city-orchestration.pl) |
 | [`riemann-hypothesis.pl`](../examples/eyelang/riemann-hypothesis.pl) | Checks a finite catalogue of non-trivial zeta zeros against the Riemann-hypothesis condition. | [`output/riemann-hypothesis.pl`](../examples/eyelang/output/riemann-hypothesis.pl) |
 | [`sat-dpll.pl`](../examples/eyelang/sat-dpll.pl) | Solves a finite SAT instance. | [`output/sat-dpll.pl`](../examples/eyelang/output/sat-dpll.pl) |
 | [`security-incident-correlation.pl`](../examples/eyelang/security-incident-correlation.pl) | Correlates security incidents across signals. | [`output/security-incident-correlation.pl`](../examples/eyelang/output/security-incident-correlation.pl) |

package/docs/eyelang-language-reference.md

@@ -39,7 +39,7 @@
   - [9.6 Strings and atom constants](#96-strings-and-atom-constants)
   - [9.7 Lists](#97-lists)
   - [9.8 Aggregation and ordering](#98-aggregation-and-ordering)
-  - [9.9 Formula terms](#99-formula-terms)
+  - [9.9 Context terms](#99-context-terms)
   - [9.10 Search control](#910-search-control)
 - [10. Extension built-ins](#10-extension-built-ins)
 - [11. Declarations](#11-declarations)
@@ -67,7 +67,7 @@
 
 eyelang is a compact Prolog-like definite-clause language for rule-based programs over ordinary terms, lists, arithmetic, strings, and finite search. An eyelang program is a finite sequence of facts and Horn clauses. The underlying declarative semantics of the pure language is **Herbrand semantics**: constants, compound terms, and lists denote themselves, and predicates denote sets of ground atomic formulas over those terms. Evaluation is goal-directed: goals are solved by unification against facts, rules, and a fixed set of built-in predicates.
 
-eyelang is intentionally smaller than ISO Prolog. It supports enough Prolog syntax to express Horn-clause reasoning, list processing, arithmetic examples, finite search, and formula data, without operators, cut, modules, dynamic predicates, DCGs, or a complete ISO standard library.
+eyelang is intentionally smaller than ISO Prolog. It supports enough Prolog syntax to express Horn-clause reasoning, list processing, arithmetic examples, finite search, and context data, without operators, cut, modules, dynamic predicates, DCGs, or a complete ISO standard library.
 
 ## 1. Terminology and normative language
 
@@ -96,7 +96,7 @@ eyelang is designed to be:
 - small enough to embed and audit;
 - deterministic in textual output order after duplicate suppression;
 - useful for relation-style `p(S, O)` output through ordinary predicate names;
-- practical for examples involving recursion, lists, arithmetic, strings, aggregation, finite search, and formula-valued data.
+- practical for examples involving recursion, lists, arithmetic, strings, aggregation, finite search, and context-valued data.
 
 Non-goals include complete ISO Prolog compatibility, operator declarations, module systems, dynamic database updates, cut-based control, and full bottom-up closure semantics.
 
@@ -252,7 +252,7 @@ Parenthesized comma terms may be goals or data:
 (name(alice, "Alice"), knows(alice, bob))
 ```
 
-When a comma term appears as a goal, it is evaluated as conjunction. When it appears as data, it remains a term. `formula_binary/4` enumerates binary members inside such formula terms.
+When a comma term appears as a goal, it is evaluated as conjunction. When it appears as data, it remains a term. `holds/2` enumerates member terms inside such contexts, and `holds/3` exposes each member as a predicate name plus an argument list for any arity.
 
 ## 6. Clauses and predicates
 
@@ -429,21 +429,23 @@ Comparisons interpret numeric-looking terms numerically. Other scalar terms are
 | `aggregate_max(Key, Template, Goal, BestKey, BestTemplate)` | Selects the solution of `Goal` with the largest resolved `Key`, returning that key and the corresponding resolved `Template`. Fails when `Goal` has no solutions. |
 | `sort(Input, Output)` | Sorts and deduplicates a proper list. |
 
-### 9.9 Formula terms
+### 9.9 Context terms
 
-Formula terms are data representations of atomic formulas and comma conjunctions.
+Context terms are data representations of atomic formulas and comma conjunctions.
 
 | Built-in | Meaning |
 |---|---|
-| `formula_binary(Formula, S, P, O)` | Enumerates binary formula members `P(S, O)`, exposing the functor as atom constant `P`. |
+| `holds(Context, Term)` | Enumerates member terms inside a context term and unifies each member with `Term`. |
+| `holds(Context, Name, Args)` | Enumerates context members of any arity, exposing each member as atom constant `Name` plus a proper argument list `Args`. |
 
 Example:
 
 ```prolog
-formula_binary((name(alice, "Alice"), knows(alice, bob)), S, P, O).
+holds((name(alice, "Alice"), knows(alice, bob)), name(S, O)).
+holds((ready, name(alice, "Alice"), route(alice, bob, 7)), Name, Args).
 ```
 
-This can yield `formula_binary((name(alice, "Alice"), knows(alice, bob)), alice, name, "Alice").` and `formula_binary((name(alice, "Alice"), knows(alice, bob)), alice, knows, bob).`
+The first goal can yield `holds((name(alice, "Alice"), knows(alice, bob)), name(alice, "Alice")).` The second can yield `holds((ready, name(alice, "Alice"), route(alice, bob, 7)), ready, []).`, `holds((ready, name(alice, "Alice"), route(alice, bob, 7)), name, [alice, "Alice"]).`, and `holds((ready, name(alice, "Alice"), route(alice, bob, 7)), route, [alice, bob, 7]).`
 
 ### 9.10 Search control
 
@@ -466,7 +468,7 @@ An extension built-in SHOULD obey the same surface-language discipline as standa
 - it SHOULD document its intended modes, especially which arguments must be ground before it runs deterministically;
 - it MUST NOT change the meaning of ordinary facts, rules, unification, or standard built-ins.
 
-For example, an implementation may include extension modules for portfolio selection, number-theory algorithms, graph search, matrix operations, or alphametic puzzles. Those modules may be valuable and may make example programs much faster, but their predicate names, arities, algorithms, and modes are implementation-defined unless they are separately standardized.
+For example, an implementation may include extension modules for number-theory algorithms, graph search, matrix operations, or alphametic puzzles. Those modules may be valuable and may make example programs much faster, but their predicate names, arities, algorithms, and modes are implementation-defined unless they are separately standardized.
 
 An implementation that provides explanation output SHOULD make extension built-ins explainable at least as opaque successful or failed built-in calls, so that proof traces do not incorrectly report "no clauses" for a host-provided relation.
 

package/examples/eyelang/annotation.pl

@@ -1,7 +1,7 @@
 % Annotation with quoted formula data.
 %
 % The program keeps the annotation as data and derives visible relations from it.
-% Formula members become default output only when explicit rules project them.
+% Context members become default output only when explicit rules project them.
 
 % Output declarations: materialize/2 selects the relations written to this example's golden output.
 materialize(name, 2).
@@ -18,17 +18,17 @@ annotation(t, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 name(S, O) :-
-  annotation(_T, Formula),
-  formula_binary(Formula, S, name, O).
+  annotation(_T, Context),
+  holds(Context, name(S, O)).
 
 log_nameOf(T, name(S, O)) :-
-  annotation(T, Formula),
-  formula_binary(Formula, S, name, O).
+  annotation(T, Context),
+  holds(Context, name(S, O)).
 
 statedBy(S, O) :-
-  annotation(_T, Formula),
-  formula_binary(Formula, S, statedBy, O).
+  annotation(_T, Context),
+  holds(Context, statedBy(S, O)).
 
 recorded(S, O) :-
-  annotation(_T, Formula),
-  formula_binary(Formula, S, recorded, O).
+  annotation(_T, Context),
+  holds(Context, recorded(S, O)).

package/examples/eyelang/context-association.pl

@@ -39,9 +39,9 @@ log_nameOf(g3, (
 % A tiny projection shows how a program can inspect a quoted context without
 % making the entire context globally true.
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
-context_statement(Context, Subject, Predicate, Object) :-
-  log_nameOf(Context, Formula),
-  formula_binary(Formula, Subject, Predicate, Object).
+context_statement(ContextName, Subject, Predicate, Object) :-
+  log_nameOf(ContextName, Context),
+  holds(Context, Predicate, [Subject, Object]).
 
 dataGraph(association, skolem_g0) :-
   context_statement(skolem_g0, bob, foaf_name, "Bob").

package/examples/eyelang/delfour.pl

@@ -4,7 +4,7 @@
 % checklist facts as relation materialization.
 %
 % Static input is kept as scoped data: the case, insight, policy, envelope, and
-% signature are formula terms, while the product catalog is a list of records.
+% signature are context terms, while the product catalog is a list of records.
 % Rules project only the fields they need, avoiding global permission/prohibition
 % facts that could contradict another policy formula in the same program.
 
@@ -41,7 +41,7 @@ materialize(marketingProhibited, 2).
 materialize(filesWrittenExpected, 2).
 
 % Program structure: facts set up the scenario, and rules derive the materialized conclusions.
-% Formula-valued facts keep each input graph scoped and easy to project.
+% Context-valued facts keep each input graph scoped and easy to project.
 case_graph(delfourCaseGraph, (
   caseName(case, "delfour"),
   requestPurpose(case, "shopping_assist"),
@@ -102,11 +102,11 @@ signature_graph(delfourSignatureGraph, (
 reason_text(reasonText, "Household requires low-sugar guidance (diabetes in POD). A neutral Insight is scoped to device 'self-scanner', event 'pick_up_scanner', retailer 'Delfour', and expires soon; the policy confines use to shopping assistance.").
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
-case_statement(S, P, O) :- case_graph(delfourCaseGraph, Formula), formula_binary(Formula, S, P, O).
-insight_statement(S, P, O) :- insight_graph(delfourInsightGraph, Formula), formula_binary(Formula, S, P, O).
-policy_statement(S, P, O) :- policy_graph(delfourPolicyGraph, Formula), formula_binary(Formula, S, P, O).
-envelope_statement(S, P, O) :- envelope_graph(delfourEnvelopeGraph, Formula), formula_binary(Formula, S, P, O).
-signature_statement(S, P, O) :- signature_graph(delfourSignatureGraph, Formula), formula_binary(Formula, S, P, O).
+case_statement(S, P, O) :- case_graph(delfourCaseGraph, Context), holds(Context, P, [S, O]).
+insight_statement(S, P, O) :- insight_graph(delfourInsightGraph, Context), holds(Context, P, [S, O]).
+policy_statement(S, P, O) :- policy_graph(delfourPolicyGraph, Context), holds(Context, P, [S, O]).
+envelope_statement(S, P, O) :- envelope_graph(delfourEnvelopeGraph, Context), holds(Context, P, [S, O]).
+signature_statement(S, P, O) :- signature_graph(delfourSignatureGraph, Context), holds(Context, P, [S, O]).
 
 case_name(case, Name) :- case_statement(case, caseName, Name).
 request_purpose(case, Purpose) :- case_statement(case, requestPurpose, Purpose).

package/examples/eyelang/dijkstra-risk-path.pl

@@ -32,8 +32,8 @@ route_network(riskNetwork, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 route_segment(From, To, Raw, Risk) :-
-  route_network(riskNetwork, Formula),
-  formula_binary(Formula, From, segment, segment(To, Raw, Risk)).
+  route_network(riskNetwork, Context),
+  holds(Context, segment(From, segment(To, Raw, Risk))).
 
 candidate(pathB, [depotA, depotB, labD]).
 candidate(pathC, [depotA, depotC, labD]).

package/examples/eyelang/dijkstra.pl

@@ -24,8 +24,8 @@ weighted_graph(dijkstraGraph, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 base_link(A, B, Cost) :-
-  weighted_graph(dijkstraGraph, Formula),
-  formula_binary(Formula, A, edge, arc(B, Cost)).
+  weighted_graph(dijkstraGraph, Context),
+  holds(Context, edge(A, arc(B, Cost))).
 
 link(A, B, Cost) :- base_link(A, B, Cost).
 link(B, A, Cost) :- base_link(A, B, Cost).

package/examples/eyelang/family-cousins.pl

@@ -33,7 +33,7 @@ family_graph(familyGraph, (
 )).
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
-family_statement(S, P, O) :- family_graph(familyGraph, Formula), formula_binary(Formula, S, P, O).
+family_statement(S, P, O) :- family_graph(familyGraph, Context), holds(Context, P, [S, O]).
 
 parent(Parent, Child) :- family_statement(Parent, parent, Child).
 branch(Person, Branch) :- family_statement(Person, seedBranch, Branch).

package/examples/eyelang/gps.pl

@@ -39,12 +39,12 @@ map_graph(mapBE, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 case_statement(S, P, O) :-
-  case_graph(caseGraph, Formula),
-  formula_binary(Formula, S, P, O).
+  case_graph(caseGraph, Context),
+  holds(Context, P, [S, O]).
 
 map_description(From, To, Action, Duration, Cost, Belief, Comfort) :-
-  map_graph(mapBE, Formula),
-  formula_binary(Formula, mapBE, gps_description, description(From, true, To, Action, Duration, Cost, Belief, Comfort)).
+  map_graph(mapBE, Context),
+  holds(Context, gps_description(mapBE, description(From, true, To, Action, Duration, Cost, Belief, Comfort))).
 
 path(From, To, [Action], Duration, Cost, Belief, Comfort) :-
   map_description(From, To, Action, Duration, Cost, Belief, Comfort).

package/examples/eyelang/odrl-dpv-healthcare-risk-ranked.pl

@@ -114,8 +114,8 @@ policy_graph(policyGraphHC1, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 policy_statement(GraphName, Subject, Predicate, Object) :-
-  policy_graph(GraphName, Formula),
-  formula_binary(Formula, Subject, Predicate, Object).
+  policy_graph(GraphName, Context),
+  holds(Context, Predicate, [Subject, Object]).
 
 permission(Graph, Permission) :- policy_statement(Graph, policyHC1, odrl_permission, Permission).
 clause(Graph, Permission, Clause) :- policy_statement(Graph, Permission, clause, Clause).

package/examples/eyelang/odrl-dpv-risk-ranked.pl

@@ -98,8 +98,8 @@ policy_graph(policyGraph1, (
 
 % Derivation rules: each rule below contributes one logical step toward the displayed results.
 policy_statement(Subject, Predicate, Object) :-
-  policy_graph(_Graph, Formula),
-  formula_binary(Formula, Subject, Predicate, Object).
+  policy_graph(_Graph, Context),
+  holds(Context, Predicate, [Subject, Object]).
 
 policy(Policy, Agreement) :- policy_statement(Policy, odrl_appliesTo, Agreement).
 permission(Policy, Rule) :- policy_statement(Policy, odrl_permission, Rule).

package/examples/eyelang/proof/annotation.pl

@@ -4,15 +4,15 @@ why(
   proof(
     goal(name(a, "Alice")),
     by(rule("annotation.pl", clause(6))),
-    bindings([binding("S", a), binding("O", "Alice"), binding("_T", t), binding("Formula", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
+    bindings([binding("S", a), binding("O", "Alice"), binding("_T", t), binding("Context", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
     uses([
       proof(
         goal(annotation(t, (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))),
         by(fact("annotation.pl", clause(5)))
       ),
       proof(
-        goal(formula_binary((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), a, name, "Alice")),
-        by(builtin(formula_binary, 4))
+        goal(holds((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), name(a, "Alice"))),
+        by(builtin(holds, 2))
       )
     ])
   )
@@ -24,15 +24,15 @@ why(
   proof(
     goal(log_nameOf(t, name(a, "Alice"))),
     by(rule("annotation.pl", clause(7))),
-    bindings([binding("T", t), binding("S", a), binding("O", "Alice"), binding("Formula", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
+    bindings([binding("T", t), binding("S", a), binding("O", "Alice"), binding("Context", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
     uses([
       proof(
         goal(annotation(t, (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))),
         by(fact("annotation.pl", clause(5)))
       ),
       proof(
-        goal(formula_binary((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), a, name, "Alice")),
-        by(builtin(formula_binary, 4))
+        goal(holds((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), name(a, "Alice"))),
+        by(builtin(holds, 2))
       )
     ])
   )
@@ -44,15 +44,15 @@ why(
   proof(
     goal(statedBy(t, bob)),
     by(rule("annotation.pl", clause(8))),
-    bindings([binding("S", t), binding("O", bob), binding("_T", t), binding("Formula", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
+    bindings([binding("S", t), binding("O", bob), binding("_T", t), binding("Context", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
     uses([
       proof(
         goal(annotation(t, (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))),
         by(fact("annotation.pl", clause(5)))
       ),
       proof(
-        goal(formula_binary((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), t, statedBy, bob)),
-        by(builtin(formula_binary, 4))
+        goal(holds((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), statedBy(t, bob))),
+        by(builtin(holds, 2))
       )
     ])
   )
@@ -64,15 +64,15 @@ why(
   proof(
     goal(recorded(t, "2021-07-07")),
     by(rule("annotation.pl", clause(9))),
-    bindings([binding("S", t), binding("O", "2021-07-07"), binding("_T", t), binding("Formula", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
+    bindings([binding("S", t), binding("O", "2021-07-07"), binding("_T", t), binding("Context", (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))]),
     uses([
       proof(
         goal(annotation(t, (name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")))),
         by(fact("annotation.pl", clause(5)))
       ),
       proof(
-        goal(formula_binary((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), t, recorded, "2021-07-07")),
-        by(builtin(formula_binary, 4))
+        goal(holds((name(a, "Alice"), statedBy(t, bob), recorded(t, "2021-07-07")), recorded(t, "2021-07-07"))),
+        by(builtin(holds, 2))
       )
     ])
   )

package/lib/eyelang/builtins/context.js

@@ -0,0 +1,42 @@
+// Context builtins that treat comma conjunctions as first-class data terms.
+// These are used by examples that construct or inspect rule bodies programmatically.
+import { atom, deref, isConjunction, listFromItems, unify } from '../term.js';
+
+export const contextBuiltins = {
+  register(registry) {
+    registry.add('holds', 2, holdsTerm);
+    registry.add('holds', 3, holdsParts);
+  }
+};
+
+function* emitContextTerm(context, term, env) {
+  context = deref(context, env);
+  if (isConjunction(context)) {
+    yield* emitContextTerm(context.args[0], term, env);
+    yield* emitContextTerm(context.args[1], term, env);
+    return;
+  }
+  const next = env.clone();
+  if (unify(term, context, next)) yield next;
+}
+
+function* emitContextParts(context, name, args, env) {
+  context = deref(context, env);
+  if (isConjunction(context)) {
+    yield* emitContextParts(context.args[0], name, args, env);
+    yield* emitContextParts(context.args[1], name, args, env);
+    return;
+  }
+  if (context.type !== 'atom' && context.type !== 'compound') return;
+  const next = env.clone();
+  const argList = context.type === 'compound' ? listFromItems(context.args) : listFromItems([]);
+  if (unify(name, atom(context.name), next) && unify(args, argList, next)) yield next;
+}
+
+function* holdsTerm({ goal, env }) {
+  yield* emitContextTerm(goal.args[0], goal.args[1], env);
+}
+
+function* holdsParts({ goal, env }) {
+  yield* emitContextParts(goal.args[0], goal.args[1], goal.args[2], env);
+}

package/lib/eyelang/builtins/registry.js

@@ -5,9 +5,8 @@ import { coreBuiltins } from './core.js';
 import { stringBuiltins } from './strings.js';
 import { listBuiltins } from './lists.js';
 import { aggregationBuiltins } from './aggregation.js';
-import { formulaBuiltins } from './formula.js';
+import { contextBuiltins } from './context.js';
 import { controlBuiltins } from './control.js';
-import { portfolioBuiltins } from './portfolio.js';
 import { searchBuiltins } from './search.js';
 import { numberTheoryBuiltins } from './number-theory.js';
 import { matrixBuiltins } from './matrix.js';
@@ -38,7 +37,7 @@ export class BuiltinRegistry {
 
 export function createDefaultRegistry() {
   const registry = new BuiltinRegistry();
-  for (const mod of [coreBuiltins, arithmeticBuiltins, stringBuiltins, listBuiltins, aggregationBuiltins, formulaBuiltins, controlBuiltins, portfolioBuiltins, searchBuiltins, numberTheoryBuiltins, matrixBuiltins, alphameticBuiltins]) {
+  for (const mod of [coreBuiltins, arithmeticBuiltins, stringBuiltins, listBuiltins, aggregationBuiltins, contextBuiltins, controlBuiltins, searchBuiltins, numberTheoryBuiltins, matrixBuiltins, alphameticBuiltins]) {
     mod.register(registry);
   }
   return registry;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "eyeling",
-  "version": "1.33.6",
+  "version": "1.33.8",
   "description": "A minimal Notation3 (N3) reasoner in JavaScript.",
   "main": "./index.js",
   "keywords": [

package/test/eyelang/conformance/README.md

@@ -36,7 +36,7 @@ The runner executes materialized programs in-process through the public JavaScri
 
 `core` covers the portable core language profile from the [eyelang language reference](../../../docs/eyelang-language-reference.md): lexical syntax, facts, definite clauses, first-order terms, lists, conjunction, structured unification through user predicates, left-to-right goal-directed proof search, materialized output, and read-back printing.
 
-`extension` covers the standard built-in and host behavior exercised by the current reference implementation: arithmetic, comparison, strings, list relations, aggregation, formula-term helpers, number-theory helpers, finite-search helpers used by the example corpus, matrix helpers, `memoize/2`, `materialize/2`, and default derived output.
+`extension` covers the standard built-in and host behavior exercised by the current reference implementation: arithmetic, comparison, strings, list relations, aggregation, context-term helpers, number-theory helpers, finite-search helpers used by the example corpus, matrix helpers, `memoize/2`, `materialize/2`, and default derived output.
 
 The profile name `extension` is a test-suite grouping name. It does not mean that these cases are outside the eyelang language reference; most of them correspond to the standard built-in profile and standard host profile in the [eyelang language reference](../../../docs/eyelang-language-reference.md).
 

package/test/eyelang/conformance/cases/extension/006_holds_parts.pl

@@ -0,0 +1,4 @@
+% Reference 9.9: holds/3 exposes context members of any arity.
+context((ready, name(alice, "Alice"), route(alice, bob, 7))).
+answer(parts, exposed(Name, Args)) :- context(C), holds(C, Name, Args).
+materialize(answer, 2).

package/test/eyelang/conformance/cases/extension/026_nested_holds_parts.pl

@@ -0,0 +1,4 @@
+% Reference 9.9: holds/3 enumerates nested comma contexts.
+context(((ready, name(a, "A")), route(a, b, 7))).
+answer(parts, exposed(Name, Args)) :- context(C), holds(C, Name, Args).
+materialize(answer, 2).

package/test/eyelang/conformance/cases/extension/027_formula_member.query

@@ -0,0 +1 @@
+answer(Label, Term)

package/test/eyelang/conformance/cases/extension/027_holds_member.pl

@@ -0,0 +1,3 @@
+% Reference 9.9: holds/2 over comma context data.
+context((name(alice, "Alice"), knows(alice, bob))).
+answer(member, Term) :- context(C), holds(C, Term).

package/test/eyelang/conformance/expected/extension/006_holds_parts.out

@@ -0,0 +1,3 @@
+answer(parts, exposed(ready, [])).
+answer(parts, exposed(name, [alice, "Alice"])).
+answer(parts, exposed(route, [alice, bob, 7])).

package/test/eyelang/conformance/expected/extension/026_nested_holds_parts.out

@@ -0,0 +1,3 @@
+answer(parts, exposed(ready, [])).
+answer(parts, exposed(name, [a, "A"])).
+answer(parts, exposed(route, [a, b, 7])).

package/test/eyelang/conformance/expected/extension/027_holds_member.out

@@ -0,0 +1,2 @@
+answer(member, name(alice, "Alice")).
+answer(member, knows(alice, bob)).

package/examples/eyelang/output/project-portfolio-optimization.pl

@@ -1,6 +0,0 @@
-bestPortfolio(portfolio2026, result(137, 75, 28, [solar, hvac, training, quality, recycling, cloud, security])).
-lowRiskTarget(portfolio2026, result(126, 74, 23, [solar, battery, hvac, training, quality, cloud])).
-feasibleCount(portfolio2026, 1390).
-projectCount(portfolio2026, 12).
-totalAvailableValue(portfolio2026, 268).
-note(portfolio2026, "aggregate builtins combine with pruning and memoization to avoid sorted candidate bags").

package/examples/eyelang/output/resilient-city-orchestration.pl

@@ -1,67 +0,0 @@
-incidentSummary(civic_storm, summary("CivicStorm Alpha", "protect people, medicine, and power without using prohibited surveillance", budget(105), riskCap(60))).
-activeSignal(civic_storm, river_surge).
-activeSignal(civic_storm, hospital_power_risk).
-activeSignal(civic_storm, medicine_cold_chain_risk).
-activeSignal(civic_storm, grid_stress).
-criticalNeed(civic_storm, evacuate_hospital).
-criticalNeed(civic_storm, deliver_medicine).
-criticalNeed(civic_storm, stabilize_substation).
-criticalNeed(civic_storm, open_shelter).
-criticalNeed(civic_storm, map_flood_front).
-policyClearance(medical_transport, deploy).
-policyClearance(cold_chain, deploy).
-policyClearance(grid_repair, deploy).
-policyClearance(shelter_ops, deploy).
-policyClearance(drone_mapping, deploy).
-policyClearance(public_info, deploy).
-usableRoute(hospital_quay, route([base, north_gate, hospital_quay], minutes(22), risk(9))).
-usableRoute(clinic_east, route([base, cold_depot, clinic_east], minutes(19), risk(5))).
-usableRoute(substation_east, route([base, north_gate, substation_east], minutes(31), risk(10))).
-usableRoute(high_school, route([base, civic_center, high_school], minutes(17), risk(4))).
-usableRoute(riverside, route([base, civic_center, riverside], minutes(21), risk(6))).
-usableRoute(civic_square, route([base, civic_center, civic_square], minutes(15), risk(3))).
-eligibleAssignment(evacuate_hospital, assignment(alpha_ambulance, [base, north_gate, hospital_quay])).
-eligibleAssignment(deliver_insulin, assignment(beta_coldchain, [base, cold_depot, clinic_east])).
-eligibleAssignment(stabilize_substation, assignment(delta_grid, [base, north_gate, substation_east])).
-eligibleAssignment(open_school_shelter, assignment(gamma_shelter, [base, civic_center, high_school])).
-eligibleAssignment(map_flood_front, assignment(echo_drone, [base, civic_center, riverside])).
-eligibleAssignment(public_dashboard, assignment(foxtrot_comms, [base, civic_center, civic_square])).
-candidateMission(evacuate_hospital, score(95, 34, 14, alpha_ambulance, [base, north_gate, hospital_quay])).
-candidateMission(deliver_insulin, score(70, 16, 9, beta_coldchain, [base, cold_depot, clinic_east])).
-candidateMission(stabilize_substation, score(85, 25, 18, delta_grid, [base, north_gate, substation_east])).
-candidateMission(open_school_shelter, score(60, 14, 6, gamma_shelter, [base, civic_center, high_school])).
-candidateMission(map_flood_front, score(45, 8, 8, echo_drone, [base, civic_center, riverside])).
-candidateMission(public_dashboard, score(28, 12, 4, foxtrot_comms, [base, civic_center, civic_square])).
-blockedMission(face_scan_crowds, policy_prohibition).
-blockedMission(resupply_west_bank, route_risk_too_high).
-bestResponse(civic_storm, response(value(355), cost(97), risk(55), missions([evacuate_hospital, deliver_insulin, stabilize_substation, open_school_shelter, map_flood_front]))).
-selectedMission(civic_storm, evacuate_hospital).
-selectedMission(civic_storm, deliver_insulin).
-selectedMission(civic_storm, stabilize_substation).
-selectedMission(civic_storm, open_school_shelter).
-selectedMission(civic_storm, map_flood_front).
-selectedTeam(evacuate_hospital, alpha_ambulance).
-selectedTeam(deliver_insulin, beta_coldchain).
-selectedTeam(stabilize_substation, delta_grid).
-selectedTeam(open_school_shelter, gamma_shelter).
-selectedTeam(map_flood_front, echo_drone).
-selectedRoute(evacuate_hospital, [base, north_gate, hospital_quay]).
-selectedRoute(deliver_insulin, [base, cold_depot, clinic_east]).
-selectedRoute(stabilize_substation, [base, north_gate, substation_east]).
-selectedRoute(open_school_shelter, [base, civic_center, high_school]).
-selectedRoute(map_flood_front, [base, civic_center, riverside]).
-coveredNeed(civic_storm, evacuate_hospital).
-coveredNeed(civic_storm, deliver_medicine).
-coveredNeed(civic_storm, stabilize_substation).
-coveredNeed(civic_storm, open_shelter).
-coveredNeed(civic_storm, map_flood_front).
-unmetNeed(civic_storm, none).
-readinessCheck(civic_storm, all_critical_needs_covered).
-readinessCheck(civic_storm, budget_and_risk_caps_respected).
-readinessCheck(civic_storm, prohibited_surveillance_rejected).
-readinessCheck(civic_storm, dangerous_west_bank_route_rejected).
-auditTrail(civic_storm, "raw sensor thresholds derive operational needs before search").
-auditTrail(civic_storm, "policy formula data is projected into deployable skill clearances").
-auditTrail(civic_storm, "bounded route search and team capabilities create candidate missions").
-auditTrail(civic_storm, "bounded_subset and aggregate_max choose the highest-value safe portfolio").
-recommendation(civic_storm, "activate the selected mission portfolio; do not deploy face recognition; defer west-bank resupply until route risk drops").

package/examples/eyelang/project-portfolio-optimization.pl

@@ -1,101 +0,0 @@
-% Capital project portfolio optimization.
-%
-% This is a practical finite-search benchmark: enumerate project portfolios
-% under budget and risk caps, pruning infeasible branches as cost and risk
-% accumulate, then use aggregate builtins to select useful optima without
-% building and sorting a large bag of candidate portfolios.
-
-% Output declarations: materialize/2 selects the relations written to this example's golden output.
-materialize(bestPortfolio, 2).
-materialize(lowRiskTarget, 2).
-materialize(feasibleCount, 2).
-materialize(projectCount, 2).
-materialize(totalAvailableValue, 2).
-materialize(note, 2).
-
-% Program structure: facts set up the scenario, and rules derive the materialized conclusions.
-budget(portfolio2026, 75).
-riskCap(portfolio2026, 28).
-targetValue(portfolio2026, 125).
-
-allProjectData([
-  p(solar, 22, 15, 4),
-  p(battery, 28, 20, 5),
-  p(hvac, 20, 13, 3),
-  p(analytics, 18, 8, 7),
-  p(training, 16, 6, 2),
-  p(robotics, 35, 30, 9),
-  p(iot, 24, 12, 8),
-  p(quality, 19, 9, 4),
-  p(resilience, 26, 18, 6),
-  p(recycling, 14, 5, 3),
-  p(cloud, 21, 11, 5),
-  p(security, 25, 16, 7)
-]).
-
-% Derivation rules: each rule below contributes one logical step toward the displayed results.
-project(Name, Value, Cost, Risk) :-
-  allProjectData(Projects),
-  member(p(Name, Value, Cost, Risk), Projects).
-
-% Cache this finite relation so bestPortfolio/2, lowRiskTarget/2, and
-% feasibleCount/2 do not re-enumerate the same search space independently.
-memoize(feasiblePortfolio, 4).
-
-choosePortfolio([], _BudgetLeft, _RiskLeft, [], 0, 0, 0).
-choosePortfolio([p(Name, ProjectValue, ProjectCost, ProjectRisk) | Rest], BudgetLeft, RiskLeft, [Name | Chosen], Value, Cost, Risk) :-
-  le(ProjectCost, BudgetLeft),
-  le(ProjectRisk, RiskLeft),
-  sub(BudgetLeft, ProjectCost, NextBudget),
-  sub(RiskLeft, ProjectRisk, NextRisk),
-  choosePortfolio(Rest, NextBudget, NextRisk, Chosen, RestValue, RestCost, RestRisk),
-  add(ProjectValue, RestValue, Value),
-  add(ProjectCost, RestCost, Cost),
-  add(ProjectRisk, RestRisk, Risk).
-choosePortfolio([_Project | Rest], BudgetLeft, RiskLeft, Chosen, Value, Cost, Risk) :-
-  choosePortfolio(Rest, BudgetLeft, RiskLeft, Chosen, Value, Cost, Risk).
-
-feasiblePortfolio(Selected, Value, Cost, Risk) :-
-  budget(portfolio2026, Budget),
-  riskCap(portfolio2026, Cap),
-  allProjectData(Projects),
-  bounded_subset(Projects, Budget, Cap, Selected, Value, Cost, Risk),
-  gt(Value, 0).
-
-scoredPortfolio(Selected, Value, Cost, Risk, NegCost, NegRisk) :-
-  feasiblePortfolio(Selected, Value, Cost, Risk),
-  neg(Cost, NegCost),
-  neg(Risk, NegRisk).
-
-targetPortfolio(Target, Selected, Value, Cost, Risk) :-
-  feasiblePortfolio(Selected, Value, Cost, Risk),
-  ge(Value, Target).
-
-bestPortfolio(portfolio2026, result(Value, Cost, Risk, Selected)) :-
-  aggregate_max(
-    [Value, NegCost, NegRisk, Selected],
-    result(Value, Cost, Risk, Selected),
-    scoredPortfolio(Selected, Value, Cost, Risk, NegCost, NegRisk),
-    _Key,
-    result(Value, Cost, Risk, Selected)).
-
-lowRiskTarget(portfolio2026, result(Value, Cost, Risk, Selected)) :-
-  targetValue(portfolio2026, Target),
-  aggregate_min(
-    [Risk, Cost, Selected],
-    result(Value, Cost, Risk, Selected),
-    targetPortfolio(Target, Selected, Value, Cost, Risk),
-    _Key,
-    result(Value, Cost, Risk, Selected)).
-
-feasibleCount(portfolio2026, Count) :-
-  countall(feasiblePortfolio(_Selected, _Value, _Cost, _Risk), Count).
-
-projectCount(portfolio2026, Count) :-
-  countall(project(_Name, _Value, _Cost, _Risk), Count).
-
-totalAvailableValue(portfolio2026, Total) :-
-  sumall(Value, project(_Name, Value, _Cost, _Risk), Total).
-
-note(portfolio2026, "aggregate builtins combine with pruning and memoization to avoid sorted candidate bags") :-
-  eq(ok, ok).

package/examples/eyelang/resilient-city-orchestration.pl

@@ -1,303 +0,0 @@
-% Resilient-city orchestration scenario.
-%
-% This example is intentionally richer than a single puzzle.  It combines raw
-% incident signals, policy-as-data, route search, team capabilities, negation,
-% aggregation, finite optimization, and materialized audit output.  The result is
-% a compact declarative emergency plan: facts describe the world; rules decide
-% which missions are lawful, reachable, and worth funding under cost/risk caps.
-
-% Output declarations: materialize/2 selects the relations written to this example's golden output.
-materialize(incidentSummary, 2).
-materialize(activeSignal, 2).
-materialize(criticalNeed, 2).
-materialize(policyClearance, 2).
-materialize(usableRoute, 2).
-materialize(eligibleAssignment, 2).
-materialize(candidateMission, 2).
-materialize(blockedMission, 2).
-materialize(bestResponse, 2).
-materialize(selectedMission, 2).
-materialize(selectedTeam, 2).
-materialize(selectedRoute, 2).
-materialize(coveredNeed, 2).
-materialize(unmetNeed, 2).
-materialize(readinessCheck, 2).
-materialize(auditTrail, 2).
-materialize(recommendation, 2).
-
-% Program structure: facts set up the storm response, and rules derive the final plan.
-incident(civic_storm).
-incident_label(civic_storm, "CivicStorm Alpha").
-incident_goal(civic_storm, "protect people, medicine, and power without using prohibited surveillance").
-budget_cap(civic_storm, 105).
-risk_cap(civic_storm, 60).
-route_risk_cap(civic_storm, 20).
-route_minutes_cap(civic_storm, 45).
-
-% Sensor readings are deliberately low-level; rules below turn them into needs.
-reading(river_gauge_cm, 318).
-threshold(river_gauge_cm, 280).
-reading(hospital_generator_hours, 5).
-threshold_low(hospital_generator_hours, 8).
-reading(clinic_refrigerator_hours, 4).
-threshold_low(clinic_refrigerator_hours, 6).
-reading(substation_load_percent, 94).
-threshold(substation_load_percent, 85).
-
-% The deployable skills are stored as quoted formula data and projected by
-% formula_binary/4, showing how policy documents can be reasoned over as data.
-policy_bundle(response_policy, (
-  permission(medical_transport, deploy),
-  permission(cold_chain, deploy),
-  permission(grid_repair, deploy),
-  permission(shelter_ops, deploy),
-  permission(drone_mapping, deploy),
-  permission(public_info, deploy),
-  prohibition(public_face_recognition, deploy)
-)).
-
-% Teams can only serve missions whose skill they possess and whose route is usable.
-team(alpha_ambulance).
-teamSkill(alpha_ambulance, medical_transport).
-teamStatus(alpha_ambulance, ready).
-team(beta_coldchain).
-teamSkill(beta_coldchain, cold_chain).
-teamStatus(beta_coldchain, ready).
-team(delta_grid).
-teamSkill(delta_grid, grid_repair).
-teamStatus(delta_grid, ready).
-team(gamma_shelter).
-teamSkill(gamma_shelter, shelter_ops).
-teamStatus(gamma_shelter, ready).
-team(echo_drone).
-teamSkill(echo_drone, drone_mapping).
-teamStatus(echo_drone, ready).
-team(foxtrot_comms).
-teamSkill(foxtrot_comms, public_info).
-teamStatus(foxtrot_comms, ready).
-team(omega_vision).
-teamSkill(omega_vision, public_face_recognition).
-teamStatus(omega_vision, ready).
-
-% The road graph is directed.  bounded_path/5 discovers simple bounded routes.
-road(base, north_gate).
-road(north_gate, hospital_quay).
-road(base, cold_depot).
-road(cold_depot, clinic_east).
-road(north_gate, substation_east).
-road(base, civic_center).
-road(civic_center, high_school).
-road(civic_center, riverside).
-road(civic_center, civic_square).
-road(base, west_bridge).
-road(west_bridge, west_bank).
-
-zone(hospital_quay).
-zone(clinic_east).
-zone(substation_east).
-zone(high_school).
-zone(riverside).
-zone(civic_square).
-zone(west_bank).
-
-routeRisk(hospital_quay, 9).
-routeRisk(clinic_east, 5).
-routeRisk(substation_east, 10).
-routeRisk(high_school, 4).
-routeRisk(riverside, 6).
-routeRisk(civic_square, 3).
-routeRisk(west_bank, 24).
-routeMinutes(hospital_quay, 22).
-routeMinutes(clinic_east, 19).
-routeMinutes(substation_east, 31).
-routeMinutes(high_school, 17).
-routeMinutes(riverside, 21).
-routeMinutes(civic_square, 15).
-routeMinutes(west_bank, 42).
-
-% Mission facts carry the operational need, location, required skill, value,
-% budget cost, and base risk.  Route risk is added by candidateMission/2.
-missionBase(evacuate_hospital, evacuate_hospital, hospital_quay, medical_transport, 95, 34, 5).
-missionBase(deliver_insulin, deliver_medicine, clinic_east, cold_chain, 70, 16, 4).
-missionBase(stabilize_substation, stabilize_substation, substation_east, grid_repair, 85, 25, 8).
-missionBase(open_school_shelter, open_shelter, high_school, shelter_ops, 60, 14, 2).
-missionBase(map_flood_front, map_flood_front, riverside, drone_mapping, 45, 8, 2).
-missionBase(public_dashboard, public_information, civic_square, public_info, 28, 12, 1).
-missionBase(face_scan_crowds, situational_awareness, civic_square, public_face_recognition, 90, 20, 2).
-missionBase(resupply_west_bank, shelter_supply, west_bank, shelter_ops, 50, 20, 5).
-
-% Memoized derived predicates keep the example readable while avoiding repeated finite searches.
-memoize(usableRoute, 2).
-memoize(candidateMission, 2).
-memoize(bestResponse, 2).
-
-% Derivation rules: each rule below contributes one logical step toward the displayed results.
-incidentSummary(Incident, summary(Label, Goal, budget(Budget), riskCap(RiskCap))) :-
-  incident_label(Incident, Label),
-  incident_goal(Incident, Goal),
-  budget_cap(Incident, Budget),
-  risk_cap(Incident, RiskCap).
-
-activeSignal(civic_storm, river_surge) :-
-  reading(river_gauge_cm, Value),
-  threshold(river_gauge_cm, Threshold),
-  ge(Value, Threshold).
-activeSignal(civic_storm, hospital_power_risk) :-
-  reading(hospital_generator_hours, Value),
-  threshold_low(hospital_generator_hours, Threshold),
-  le(Value, Threshold).
-activeSignal(civic_storm, medicine_cold_chain_risk) :-
-  reading(clinic_refrigerator_hours, Value),
-  threshold_low(clinic_refrigerator_hours, Threshold),
-  le(Value, Threshold).
-activeSignal(civic_storm, grid_stress) :-
-  reading(substation_load_percent, Value),
-  threshold(substation_load_percent, Threshold),
-  ge(Value, Threshold).
-
-criticalNeed(civic_storm, evacuate_hospital) :-
-  activeSignal(civic_storm, river_surge),
-  activeSignal(civic_storm, hospital_power_risk).
-criticalNeed(civic_storm, deliver_medicine) :-
-  activeSignal(civic_storm, medicine_cold_chain_risk).
-criticalNeed(civic_storm, stabilize_substation) :-
-  activeSignal(civic_storm, grid_stress).
-criticalNeed(civic_storm, open_shelter) :-
-  activeSignal(civic_storm, river_surge).
-criticalNeed(civic_storm, map_flood_front) :-
-  activeSignal(civic_storm, river_surge).
-
-policyClearance(Skill, deploy) :-
-  policy_bundle(response_policy, Formula),
-  formula_binary(Formula, Skill, permission, deploy),
-  not(policyBlocked(Skill, deploy)).
-
-policyBlocked(Skill, deploy) :-
-  policy_bundle(response_policy, Formula),
-  formula_binary(Formula, Skill, prohibition, deploy).
-
-usableRoute(Zone, route(Path, minutes(Minutes), risk(RouteRisk))) :-
-  zone(Zone),
-  bounded_path(road, base, Zone, 3, Path),
-  routeMinutes(Zone, Minutes),
-  routeRisk(Zone, RouteRisk),
-  route_minutes_cap(civic_storm, MinutesCap),
-  route_risk_cap(civic_storm, RiskCap),
-  le(Minutes, MinutesCap),
-  le(RouteRisk, RiskCap).
-
-eligibleAssignment(Mission, assignment(Team, Path)) :-
-  missionBase(Mission, _Need, Zone, Skill, _Value, _Cost, _BaseRisk),
-  policyClearance(Skill, deploy),
-  teamSkill(Team, Skill),
-  teamStatus(Team, ready),
-  usableRoute(Zone, route(Path, minutes(_Minutes), risk(_RouteRisk))).
-
-candidateMission(Mission, score(Value, Cost, TotalRisk, Team, Path)) :-
-  missionBase(Mission, _Need, Zone, Skill, Value, Cost, BaseRisk),
-  policyClearance(Skill, deploy),
-  teamSkill(Team, Skill),
-  teamStatus(Team, ready),
-  usableRoute(Zone, route(Path, minutes(_Minutes), risk(RouteRisk))),
-  add(BaseRisk, RouteRisk, TotalRisk).
-
-blockedMission(Mission, policy_prohibition) :-
-  missionBase(Mission, _Need, _Zone, Skill, _Value, _Cost, _BaseRisk),
-  policyBlocked(Skill, deploy).
-blockedMission(Mission, route_risk_too_high) :-
-  missionBase(Mission, _Need, Zone, _Skill, _Value, _Cost, _BaseRisk),
-  routeRisk(Zone, RouteRisk),
-  route_risk_cap(civic_storm, Cap),
-  gt(RouteRisk, Cap).
-blockedMission(Mission, no_eligible_assignment) :-
-  missionBase(Mission, _Need, _Zone, _Skill, _Value, _Cost, _BaseRisk),
-  not(candidateMission(Mission, _Score)),
-  not(blockedMission(Mission, policy_prohibition)),
-  not(blockedMission(Mission, route_risk_too_high)).
-
-missionItem(p(Mission, Value, Cost, Risk)) :-
-  candidateMission(Mission, score(Value, Cost, Risk, _Team, _Path)).
-
-missionItems(Items) :-
-  findall(Item, missionItem(Item), Items).
-
-missionAddressesNeed(Mission, Need) :-
-  missionBase(Mission, Need, _Zone, _Skill, _Value, _Cost, _BaseRisk).
-
-planCoversNeed(Plan, Need) :-
-  member(Mission, Plan),
-  missionAddressesNeed(Mission, Need).
-
-uncoveredNeedInPlan(Plan, Need) :-
-  criticalNeed(civic_storm, Need),
-  not(planCoversNeed(Plan, Need)).
-
-coversAllCriticalNeeds(Plan) :-
-  not(uncoveredNeedInPlan(Plan, _Need)).
-
-candidateResponse(Plan, Value, Cost, Risk, NegCost, NegRisk) :-
-  budget_cap(civic_storm, Budget),
-  risk_cap(civic_storm, RiskCap),
-  missionItems(Items),
-  bounded_subset(Items, Budget, RiskCap, Plan, Value, Cost, Risk),
-  coversAllCriticalNeeds(Plan),
-  neg(Cost, NegCost),
-  neg(Risk, NegRisk).
-
-bestResponse(civic_storm, response(value(Value), cost(Cost), risk(Risk), missions(Plan))) :-
-  aggregate_max(
-    [Value, NegCost, NegRisk, Plan],
-    response(value(Value), cost(Cost), risk(Risk), missions(Plan)),
-    candidateResponse(Plan, Value, Cost, Risk, NegCost, NegRisk),
-    _Key,
-    response(value(Value), cost(Cost), risk(Risk), missions(Plan))).
-
-selectedMission(civic_storm, Mission) :-
-  bestResponse(civic_storm, response(value(_Value), cost(_Cost), risk(_Risk), missions(Plan))),
-  member(Mission, Plan).
-
-selectedTeam(Mission, Team) :-
-  selectedMission(civic_storm, Mission),
-  candidateMission(Mission, score(_Value, _Cost, _Risk, Team, _Path)).
-
-selectedRoute(Mission, Path) :-
-  selectedMission(civic_storm, Mission),
-  candidateMission(Mission, score(_Value, _Cost, _Risk, _Team, Path)).
-
-coveredNeed(civic_storm, Need) :-
-  selectedMission(civic_storm, Mission),
-  missionAddressesNeed(Mission, Need).
-
-unmetNeed(civic_storm, Need) :-
-  bestResponse(civic_storm, response(value(_Value), cost(_Cost), risk(_Risk), missions(Plan))),
-  uncoveredNeedInPlan(Plan, Need).
-unmetNeed(civic_storm, none) :-
-  bestResponse(civic_storm, response(value(_Value), cost(_Cost), risk(_Risk), missions(Plan))),
-  coversAllCriticalNeeds(Plan).
-
-readinessCheck(civic_storm, all_critical_needs_covered) :-
-  unmetNeed(civic_storm, none).
-readinessCheck(civic_storm, budget_and_risk_caps_respected) :-
-  bestResponse(civic_storm, response(value(_Value), cost(Cost), risk(Risk), missions(_Plan))),
-  budget_cap(civic_storm, Budget),
-  risk_cap(civic_storm, Cap),
-  le(Cost, Budget),
-  le(Risk, Cap).
-readinessCheck(civic_storm, prohibited_surveillance_rejected) :-
-  blockedMission(face_scan_crowds, policy_prohibition).
-readinessCheck(civic_storm, dangerous_west_bank_route_rejected) :-
-  blockedMission(resupply_west_bank, route_risk_too_high).
-
-auditTrail(civic_storm, "raw sensor thresholds derive operational needs before search") :-
-  eq(ok, ok).
-auditTrail(civic_storm, "policy formula data is projected into deployable skill clearances") :-
-  eq(ok, ok).
-auditTrail(civic_storm, "bounded route search and team capabilities create candidate missions") :-
-  eq(ok, ok).
-auditTrail(civic_storm, "bounded_subset and aggregate_max choose the highest-value safe portfolio") :-
-  eq(ok, ok).
-
-recommendation(civic_storm, "activate the selected mission portfolio; do not deploy face recognition; defer west-bank resupply until route risk drops") :-
-  readinessCheck(civic_storm, all_critical_needs_covered),
-  readinessCheck(civic_storm, prohibited_surveillance_rejected),
-  readinessCheck(civic_storm, dangerous_west_bank_route_rejected).

package/lib/eyelang/builtins/formula.js

@@ -1,26 +0,0 @@
-// Formula builtins that treat conjunctions as first-class data terms.
-// These are used by examples that construct or inspect rule bodies programmatically.
-import { atom, deref, isConjunction, unify } from '../term.js';
-
-export const formulaBuiltins = {
-  register(registry) {
-    registry.add('formula_binary', 4, formulaBinary);
-  }
-};
-
-
-function* emitFormulaBinary(formula, subject, predicate, object, env) {
-  formula = deref(formula, env);
-  if (isConjunction(formula)) {
-    yield* emitFormulaBinary(formula.args[0], subject, predicate, object, env);
-    yield* emitFormulaBinary(formula.args[1], subject, predicate, object, env);
-    return;
-  }
-  if (formula.type !== 'compound' || formula.arity !== 2) return;
-  const next = env.clone();
-  if (unify(subject, formula.args[0], next) && unify(predicate, atom(formula.name), next) && unify(object, formula.args[1], next)) yield next;
-}
-
-function* formulaBinary({ goal, env }) {
-  yield* emitFormulaBinary(goal.args[0], goal.args[1], goal.args[2], goal.args[3], env);
-}

package/lib/eyelang/builtins/portfolio.js

@@ -1,73 +0,0 @@
-// Reusable bounded-subset optimizer for finite knapsack-style examples.
-// Items are p(Name, Value, Cost, Risk) terms; the builtin enumerates every
-// subset whose total cost/risk stays within the supplied caps.
-import { deref, listFromItems, numberTerm, properListItems, unify } from '../term.js';
-
-export const portfolioBuiltins = {
-  register(registry) {
-    registry.add('bounded_subset', 7, boundedSubset, {
-      fallbackWhenNotReady: true,
-      ready: boundedSubsetReady,
-    });
-  }
-};
-
-function boundedSubsetReady(goal, env) {
-  return parseItems(goal.args[0], env) !== null && intArg(goal.args[1], env) !== null && intArg(goal.args[2], env) !== null;
-}
-
-function* boundedSubset({ goal, env }) {
-  const items = parseItems(goal.args[0], env);
-  const budget = intArg(goal.args[1], env);
-  const riskCap = intArg(goal.args[2], env);
-  if (!items || budget == null || riskCap == null) return;
-  for (const answer of enumerate(items, 0, budget, riskCap)) {
-    const next = env.clone();
-    if (unify(goal.args[3], listFromItems(answer.names), next) &&
-        unify(goal.args[4], numberTerm(answer.value), next) &&
-        unify(goal.args[5], numberTerm(answer.cost), next) &&
-        unify(goal.args[6], numberTerm(answer.risk), next)) {
-      yield next;
-    }
-  }
-}
-
-function* enumerate(items, index, budget, riskLeft) {
-  if (index >= items.length) {
-    yield { names: [], value: 0, cost: 0, risk: 0 };
-    return;
-  }
-  const item = items[index];
-  if (item.cost <= budget && item.risk <= riskLeft) {
-    for (const rest of enumerate(items, index + 1, budget - item.cost, riskLeft - item.risk)) {
-      yield {
-        names: [item.name, ...rest.names],
-        value: item.value + rest.value,
-        cost: item.cost + rest.cost,
-        risk: item.risk + rest.risk,
-      };
-    }
-  }
-  yield* enumerate(items, index + 1, budget, riskLeft);
-}
-
-function parseItems(term, env) {
-  const list = properListItems(term, env);
-  if (!list) return null;
-  const items = [];
-  for (const itemTerm of list) {
-    const item = deref(itemTerm, env);
-    if (item.type !== 'compound' || item.name !== 'p' || item.arity !== 4) return null;
-    const value = intArg(item.args[1], env), cost = intArg(item.args[2], env), risk = intArg(item.args[3], env);
-    if (value == null || cost == null || risk == null) return null;
-    items.push({ name: item.args[0], value, cost, risk });
-  }
-  return items;
-}
-
-function intArg(term, env) {
-  const value = deref(term, env);
-  if (value.type !== 'number' || !/^-?\d+$/.test(value.name)) return null;
-  const n = Number(value.name);
-  return Number.isSafeInteger(n) ? n : null;
-}

package/test/eyelang/conformance/cases/extension/006_formula_terms.pl

@@ -1,4 +0,0 @@
-% Reference 9.9: formula_binary/4 over comma formula data.
-formula((name(alice, "Alice"), knows(alice, bob))).
-answer(binary, exposed(S, P, O)) :- formula(F), formula_binary(F, S, P, O).
-materialize(answer, 2).

package/test/eyelang/conformance/cases/extension/026_nested_formula_terms.pl

@@ -1,4 +0,0 @@
-% Reference 9.9: formula helpers enumerate nested comma formula members.
-formula(((name(a, "A"), knows(a, b)), likes(b, c))).
-answer(binary, exposed(S, P, O)) :- formula(F), formula_binary(F, S, P, O).
-materialize(answer, 2).

package/test/eyelang/conformance/cases/extension/046_bounded_subset.pl

@@ -1,4 +0,0 @@
-% Reference 9.5: bounded_subset/7 enumerates subsets within budget and risk caps.
-projects([p(a, 5, 3, 1), p(b, 4, 2, 2), p(c, 7, 5, 3)]).
-answer(selection, result(Names, Value, Cost, Risk)) :- projects(Ps), bounded_subset(Ps, 5, 3, Names, Value, Cost, Risk).
-materialize(answer, 2).

package/test/eyelang/conformance/expected/extension/006_formula_terms.out

@@ -1,2 +0,0 @@
-answer(binary, exposed(alice, name, "Alice")).
-answer(binary, exposed(alice, knows, bob)).

package/test/eyelang/conformance/expected/extension/026_nested_formula_terms.out

@@ -1,3 +0,0 @@
-answer(binary, exposed(a, name, "A")).
-answer(binary, exposed(a, knows, b)).
-answer(binary, exposed(b, likes, c)).

package/test/eyelang/conformance/expected/extension/046_bounded_subset.out

@@ -1,5 +0,0 @@
-answer(selection, result([a, b], 9, 5, 3)).
-answer(selection, result([a], 5, 3, 1)).
-answer(selection, result([b], 4, 2, 2)).
-answer(selection, result([c], 7, 5, 3)).
-answer(selection, result([], 0, 0, 0)).