joyful-ii-V-I · Oct 7, 2026 · Oct 1, 2026 · Oct 1, 2026 · Oct 1, 2026 · Oct 1, 2026
diff --git a/CHANGELOG.md b/CHANGELOG.md
diff --git a/README.md b/README.md
@@ -2082,9 +2082,9 @@ wrong, and it has. These are the results that say so, all in-tree, all published
 ### In the tests
 
 <details>
-<summary><b>664 gate scripts</b>, five contracts no unit test can hold, and the house rule: write the gate before the code it measures</summary> <!-- gatecount -->
+<summary><b>671 gate scripts</b>, five contracts no unit test can hold, and the house rule: write the gate before the code it measures</summary> <!-- gatecount -->
 
-`test/regression.sh` names **664 gate scripts** and is the authoritative list; <!-- gatecount -->
+`test/regression.sh` names **671 gate scripts** and is the authoritative list; <!-- gatecount -->
 `python3 test/pargates.py . ./build/ripwire -j 6` runs the same set in parallel. On top of them sit the
 contracts that do not fit a unit test: two runs byte-identical, warm output identical to cold, output
 that pipes clean through `xmllint --noout`, a sanitizer build with `-fno-sanitize-recover=all`, and a

diff --git a/docs/COMMANDS.md b/docs/COMMANDS.md
@@ -3350,11 +3350,11 @@ $ ./build/ripwire . --slice-depth=3
 _Hold a LOCATION, not a name: the enclosing-definition chain at FILE:LINE (a compiler error, a diff hunk, a stack frame), outermost -> innermost._
 
 ```
-$ ./build/ripwire . --at=src/graph.h:5565
+$ ./build/ripwire . --at=src/graph.h:7397
 <!-- ripwire at: the ENCLOSING-DEFINITION CHAIN at one FILE:LINE seed. p= the resolved file, l= the 1-based seed line, sym= the innermost enclosing definition's name (what the same seed resolves to in a selector position), chain= the row count. Rows are INDEXED definitions only, outermost first, innermost last: n= the definition's name, t= its kind tag, l= its own start line, el= its end line (1-based, inclusive). A namespace or any construct the index does not carry is NOT a row, so an outer scope can be absent rather than misnamed; a seed line inside no indexed definition is refused, never served as an empty chain. The same seed composes into any SYM selector as @FILE:LINE (callers, callees, impact, around, expand, uses, edit-check, slice, safe-delete, path, connect) and resolves to the innermost row. -->
 <!-- root= on this element is the crawl root every p= below is RELATIVE to (single-root runs only; absent => p= is the path ingest itself used, unchanged). -->
-<at p="src/graph.h" l="5565" sym="rankGraphTeleport" chain="1" root=".">
-<s n="rankGraphTeleport" t="fn" l="5565" el="5593"/>
+<at p="src/graph.h" l="7397" sym="rankGraphTeleport" chain="1" root=".">
+<s n="rankGraphTeleport" t="fn" l="7397" el="7425"/>
 </at>
 ```
 

diff --git a/docs/EVALS.md b/docs/EVALS.md
@@ -21,7 +21,7 @@ section, and it is not an afterthought.
 | **Co-change / known-item evals** | `--eval`, `--eval-retrieval` (see `bench/ANSWERQUALITY.md`) | Whether the tool surfaces the other files a real historical commit touched; and known-item retrieval across four rankers. |
 | **Ensemble calibration harness** | `bench/ensemblecal/` | Whether `--ensemble`'s four evidence families are actually orthogonal, how often each fires, how stable each is across commits — and the preset ladder derived from that (§9). |
 | **Differential argv harness** | `test/argvdiffcheck.sh` | That a refactor changed *nothing observable*: two binaries, every argv vector, stdout + stderr + exit code byte-identical. |
-| **The gate suite** | `test/regression.sh`, `test/pargates.py` | 664 gate scripts plus the determinism, cache-transparency and golden contracts. <!-- gatecount --> |
+| **The gate suite** | `test/regression.sh`, `test/pargates.py` | 671 gate scripts plus the determinism, cache-transparency and golden contracts. <!-- gatecount --> |
 | **`--quality-delta`** | `src/quality.h` | Ten measured code-quality failure modes, reported only where a change made them worse. |
 
 ### The labeling protocol (why the held-out eval is allowed to disagree with the ranker)
@@ -5866,7 +5866,7 @@ copy here would be exactly the dialect divergence that gate exists to catch. Com
 tags, wrap, stable-order defaults), seven individually invoked standalone gates (`g1freshcheck`,
 `skillscan`, `htmlexport`, `compresscheck`, `handoffcheck`, `releaseinstallcheck`,
 `taskroutecheck`), and a single loop
-naming **664 gate scripts**, all of which exist on disk. <!-- gatecount -->
+naming **671 gate scripts**, all of which exist on disk. <!-- gatecount -->
 
 `python3 test/pargates.py . ./build/ripwire -j 6` runs the same scripts in parallel so a full
 verification fits in one sitting. It does not modify `regression.sh`.
@@ -6969,7 +6969,7 @@ Listed because the reason is more useful than the silence.
   shipped**. See `bench/locbench/anchorhop_calib.json`. The mention anchor's reproducible numbers are
   the ablations in §4.
 - **A single round gate-count.** Two in-tree numbers disagree (`test/pargates.py`'s docstring says
-  ~210; `test/argvdiffcheck.sh` says 200+), while the loop in `test/regression.sh` names 664. The <!-- gatecount -->
+  ~210; `test/argvdiffcheck.sh` says 200+), while the loop in `test/regression.sh` names 671. The <!-- gatecount -->
   loop is the authority; the stale docstrings are a known drift. Since 2026-09-10 the number is not
   written by hand anywhere: `docs/gatecount_build.py` derives it from the loop and rewrites every
   published site, `test/gatecountcheck.sh` fails if any of them drifts, and `test/manifestcheck.sh`

diff --git a/docs/captures/COMMANDS_showcase_2026-09-14.md b/docs/captures/COMMANDS_showcase_2026-09-14.md
@@ -3921,29 +3921,29 @@ ripwire 0.6.1 (dev, AppleClang 21.0.0.21000101, emit=std::print, built_from=f8d4
 
 # navigate — seeds, claims, slices, shapes
 
-## `./build/ripwire . --at=src/graph.h:5565`
+## `./build/ripwire . --at=src/graph.h:7397`
 
 *Hold a LOCATION, not a name: the enclosing-definition chain at FILE:LINE (a compiler error, a diff hunk, a stack frame), outermost -> innermost.*
 
 `````
 <!-- ripwire at: the ENCLOSING-DEFINITION CHAIN at one FILE:LINE seed. p= the resolved file, l= the 1-based seed line, sym= the innermost enclosing definition's name (what the same seed resolves to in a selector position), chain= the row count. Rows are INDEXED definitions only, outermost first, innermost last: n= the definition's name, t= its kind tag, l= its own start line, el= its end line (1-based, inclusive). A namespace or any construct the index does not carry is NOT a row, so an outer scope can be absent rather than misnamed; a seed line inside no indexed definition is refused, never served as an empty chain. The same seed composes into any SYM selector as @FILE:LINE (callers, callees, impact, around, expand, uses, edit-check, slice, safe-delete, path, connect) and resolves to the innermost row. -->
 <!-- root= on this element is the crawl root every p= below is RELATIVE to (single-root runs only; absent => p= is the path ingest itself used, unchanged). -->
-<at p="src/graph.h" l="5565" sym="rankGraphTeleport" chain="1" root=".">
-<s n="rankGraphTeleport" t="fn" l="5565" el="5593"/>
+<at p="src/graph.h" l="7397" sym="rankGraphTeleport" chain="1" root=".">
+<s n="rankGraphTeleport" t="fn" l="7397" el="7425"/>
 </at>
 `````
 
-## `./build/ripwire . --callers=@src/graph.h:5565`
+## `./build/ripwire . --callers=@src/graph.h:7397`
 
 *The same seed in a SELECTOR position: @FILE:LINE resolves to the innermost enclosing definition, then --callers runs on it.*
 
 `````
 <!-- ripwire callers/callees: the 1-hop call hierarchy read off the call graph — the callers form lists symbols that CALL of=; the callees form lists symbols of= itself calls. of= is the selector you passed, defs= how many DEFINITIONS it resolved to (rows UNION every def's neighbours), count= the DISTINCT neighbour symbols (a floor, per counts_floor=), windowed by limit= and offset=. A neighbour that is an indexed function-like #define is a macro row (t="macro", role="macro" on the XML row): the edge crosses a macro expansion, not a plain call — rows carry no role= otherwise. Rows are ordered SOURCE first, then test/bench, then docs, by path within a tier. hop_tested=/hop_untested= partition count= by the tested= lens below (1-hop, never transitive). tested="1" on a row means an indexed test transitively reaches it (never 0, omitted when it does not). BLIND SPOT the test-gate legend also names: only a CALL EDGE from an INDEXED test symbol counts here, so a shell or CLI-level test running a built binary as a SUBPROCESS is invisible to it and a repo tested that way reads all-untested. Read untested= as no in-process test reaches it, not as no test covers it. next= is the one pasteable follow-up (the uses verb on this selector: the call sites). counts_floor="1" means every count here is a FLOOR, never a total: edges are extracted from source TEXT by NAME. Missing: dynamic dispatch (virtual/interface/duck-typed), a most-vexing-parse declaration with no call expression, a function-pointer/callback bound to more than one function in scope (reassigned, table-indexed, lambda-bound, or address-taken/reference-bound), and a plain-name binding (fp=handler) whose variable type is not PROVABLY a function pointer (a same-file typedef/declarator; a HEADER typedef is missed; auto/template types are read as unpinned, so KEPT). A macro-generated call site is role="macro" only when its name uniquely names an indexed function-like #define (C-family, t="macro"); a shared name stays a plain call, an unindexed macro is no edge. Read a zero as "none found", never as "none exists". graph_ambiguous=/graph_unresolved= are the whole graph's resolver gauge (calls split over several defs / calls whose in-repo defs were all language-filtered), the map header's ambiguous=/unresolved=. graph_unindexed=N is a third gauge: files no grammar could read (the map header's unindexed=), whose calls raise neither gauge above; absent when zero, and so is this sentence. COUNTING UNIT differs by verb: callers, callees, edit-check, graph-query and pr-context counts are DISTINCT SYMBOLS (repeated calls from one caller, and calls to two overloads, collapse into ONE row; multiplicity survives only in the call graph's edge weight). The reach counts (impact's reaches=, pr-context's dependents=) are the size of a transitive reach SET, each symbol counted once. The uses verb counts call SITES, one row per occurrence — a larger count there for the same symbol is these units agreeing, not disagreeing. The map header's edges= is a unit again different — distinct (caller,callee) PAIRS — and that document carries neither this marker nor this clause. -->
 <!-- root= on this element is the crawl root every p= below is RELATIVE to (single-root runs only; absent => p= is the path ingest itself used, unchanged). -->
-<callers of="@src/graph.h:5565" defs="1" count="7" root="." hop_tested="0" hop_untested="7" graph_ambiguous="10252" graph_unresolved="12843" graph_unindexed="237" counts_floor="1" next="--uses=@src/graph.h:5565">
+<callers of="@src/graph.h:7397" defs="1" count="7" root="." hop_tested="0" hop_untested="7" graph_ambiguous="10473" graph_unresolved="12903" graph_unindexed="237" counts_floor="1" next="--uses=@src/graph.h:7397">
 <s t="fn" n="getIndex" p="src/mcpindex.h:1165"/>
-<s t="fn" n="rankGraph" p="src/graph.h:5606"/>
-<s t="fn" n="anchoredLexicalRank" p="src/graph.h:6252"/>
+<s t="fn" n="rankGraph" p="src/graph.h:7438"/>
+<s t="fn" n="anchoredLexicalRank" p="src/graph.h:8084"/>
 <s t="fn" n="runEval" p="src/eval.h:171"/>
 <s t="fn" n="churnDecayRanking" p="src/main.cpp:1380"/>
 <s t="fn" n="churnRankedGraph" p="src/main.cpp:1419"/>

diff --git a/present/deck5_ripwire_build.js b/present/deck5_ripwire_build.js
@@ -1214,7 +1214,7 @@ function storyCards(s, { kick, head, stories, footText }){
   kicker(s, "// how it stays true", AMBER);
   title(s, "Proven, not promised");
   const cards = [
-    ["664 gate scripts", "the suite runs on every push — plus determinism, cache-transparency and golden contracts; the gate count itself is gated against the runner's own loop"], // gatecount
+    ["671 gate scripts", "the suite runs on every push — plus determinism, cache-transparency and golden contracts; the gate count itself is gated against the runner's own loop"], // gatecount
     ["byte-identical, always", "two runs over the same tree produce the same bytes; warm equals cold. Enforced in CI, twice — Release AND a plain flavour, because NDEBUG once blinded a whole class of checks"],
     ["differential refactoring", "a refactor must prove it changed nothing observable: two binaries, hundreds of argv vectors, stdout + stderr + exit codes byte-identical"],
     ["held-out labels, authored blind", "eval labels were written by reading source before the ranker ever ran on them — so the eval is allowed to say the ranker is wrong. It has."],
@@ -1238,7 +1238,7 @@ function storyCards(s, { kick, head, stories, footText }){
   title(s, "Claims you can trust, because we publish what failed", { size: 32 });
 
   card(s, MX, 1.72, 3.86, 1.72);
-  stat(s, "664", "gate scripts named by test/regression.sh — and the COUNT itself is gated against the runner's own loop, so it cannot go stale quietly", // gatecount
+  stat(s, "671", "gate scripts named by test/regression.sh — and the COUNT itself is gated against the runner's own loop, so it cannot go stale quietly", // gatecount
     MX+0.15, 1.86, 3.56, CYAN, { bsize: 42, bh: 0.66, lsize: 9.5 });
   card(s, 4.68, 1.72, 3.86, 1.72, CARD2);
   stat(s, "8", "registered NEGATIVES — changes built, gated green, measured against a band written before the code, and reverted rather than tuned",
@@ -1540,7 +1540,7 @@ function storyCards(s, { kick, head, stories, footText }){
     ["185 long flags · 36 slides",        "bash test/deckclaimcheck.sh"],
     ["every --flag named here exists",    "bash test/deckcheck.sh"],
     ["74.7% fewer element bytes",         "bash test/showcasecapturecheck.sh"],
-    ["664 gate scripts",                  "bash test/manifestcheck.sh"], // gatecount
+    ["671 gate scripts",                  "bash test/manifestcheck.sh"], // gatecount
     ["49 repos · 71 papers · 237 surveyed","bash test/readmedriftcheck.sh"],
     ["the ten moments, any row",          "ripwire . --callers=SYM | wc -c"],
     ["the head-to-head table",            "bench/headtohead/r4-2026-08-06/"],

diff --git a/queries/ruby/tags.scm b/queries/ruby/tags.scm
@@ -5,9 +5,11 @@
 ; Ruby structure the call graph cares about:
 ;   - class / module definitions → def nodes (containers; module is a namespace/mixin)
 ;   - `def name` / `def self.name` method definitions → the def nodes calls resolve TO
-;   - method calls → the call references (edges). A bare `foo` with no receiver and no
-;     args parses as (identifier); `foo(..)` / `obj.foo` parses as (call). We capture the
-;     (call) form's method name — the reliable, unambiguous call site.
+;   - method calls → the call references (edges). `foo(..)` / `obj.foo` parses as (call), and
+;     this query captures the (call) form's method name. A bare `foo` with no receiver, args or
+;     parentheses parses as (identifier) — the same node a local-variable read is — and no query
+;     can tell the two apart: ingest_binds.h captureRubyBareCalls mints those calls by Ruby's own
+;     lexical local rule instead (parser version 130, test/rubybarecallcheck.sh).
 ;   - require / require_relative → reference edges (they ARE plain (call) nodes, captured
 ;     by the call rule below — no separate rule needed).
 ;

diff --git a/src/darkflags.h b/src/darkflags.h
@@ -449,7 +449,7 @@ inline constexpr std::int32_t kTemplateText = -1;
 inline constexpr std::string_view kShellExtTable[] = { ".sh", ".bash", ".zsh" };
 
 inline constexpr std::string_view kHashCommentExtTable[] = {
-    ".py", ".pyi", ".sh", ".bash", ".zsh", ".rb", ".pl", ".cmake", ".yml", ".yaml", ".toml", ".r", ".jl"
+    ".py", ".pyi", ".sh", ".bash", ".zsh", ".rb", ".rake", ".jbuilder", ".pl", ".cmake", ".yml", ".yaml", ".toml", ".r", ".jl"
 };
 
 inline LineSyntax lineSyntaxFor( std::string_view path )