diff --git a/.claude/skills/t27f/SKILL.md b/.claude/skills/t27f/SKILL.md
index eb5da0af3a..755443143e 100644
--- a/.claude/skills/t27f/SKILL.md
+++ b/.claude/skills/t27f/SKILL.md
@@ -81,7 +81,7 @@ the diagram core in detail, then the other surfaces and the topic-map applicatio
 | Spec -> graph IR schema (the core) | `specs/t27f/spec_graph.t27` (t27#10020) | spec with tests: node / edge kinds, labels, spans, `validate_graph`, line -> node mapping, node -> ui_plan widget, one hand-read fixture (`specs/tutorial/08_modules_and_arrays.t27`) |
 | Graph view: layout of a spec graph | `specs/t27f/spec_graph_view.t27` (t27#10034) | spec with tests: one column per node kind, size / colour / dash / row order from the inference label, UNKNOWN nodes alone in a band, every edge with a source line, node -> widget through `node_widget`; numbers only, laid out on the hand-read fixture; drawn by the MCP tool `t27f_graph` (gHashTag/t27-mcp#9) |
 | Graph view for big specs: wrap, fold, expand, hit | `specs/t27f/spec_graph_fold.t27` (t27#10218) | spec with tests: a kind column taller than N = 12 rows wraps into sub-columns so the canvas aspect is near phi (identical to the view when nothing wraps); a fold to one box per kind with the exact count and one arc per pair of kinds with the exact edge count, each labelled by its weakest member; `expand_graph` = one kind plus its direct neighbours, copied verbatim with an origin map; `wrap_hit` / `wrap_hit_line_start` / `wrap_hit_line_end` take a click to a node and its lines; numbers only, the view's colours are not touched |
-| Dependencies view: the specs around one spec | `specs/t27f/spec_deps.t27` (t27#10680) | spec with tests: reads `t27c use-edges --specs-dir specs`; layer 0 is the focus, +k the specs k imports away (right), -k the specs k importers away (left), each spec once in its shortest-path layer; every edge-list line between two drawn specs is one `exact` edge, a repeated line drawn once; `cycle=1` when the focus imports itself or is imported by a spec in its own import closure; a walk past the caller's room or depth is refused whole (`refused\tfull` / `refused\tdeep`), never cut. Limits: `use` lines t27c does not resolve are not drawn (#7176); cycles that miss the focus are not reported; a spec on a cycle through the focus is drawn on both sides; no host draws it yet. Corpus on master e607228f7: 806 specs, 2461 edge lines, nothing refused at depth 32, deepest walks 6, one real cycle (igla race `backend` <-> `rtl`) |
+| Dependencies view: the specs around one spec | `specs/t27f/spec_deps.t27` (t27#10680) | spec with tests: reads `t27c use-edges --specs-dir specs`; layer 0 is the focus, +k the specs k imports away (right), -k the specs k importers away (left), each spec once in its shortest-path layer; every edge-list line between two drawn specs is one `exact` edge, a repeated line drawn once; `cycle=1` when the focus imports itself or is imported by a spec in its own import closure; a walk past the caller's room or depth is refused whole (`refused\tfull` / `refused\tdeep`), never cut. Limits: `use` lines t27c does not resolve are not drawn (#7176); the view flags only the focus, and `deps_cycles` (t27#10693) lists every spec on a cycle, each once, refused whole (`cap + 1`) if anything does not fit; a spec on a cycle through the focus is drawn on both sides; no host draws it yet. Corpus on master e607228f7: 806 specs, 2461 edge lines, nothing refused at depth 32, deepest walks 6, one real cycle (igla race `backend` <-> `rtl`), which `deps_cycles` lists exactly on master f7efbc232 and lists as empty with #10689 |
 | Spec -> graph IR exporter | `specs/t27f/spec_graph_export.t27` (t27#10162) | spec with tests: `export_graph(path, src)` is a deterministic byte scanner over the source, `export_verdict` is `validate_graph` of the result. One node per top-level declaration (a leading `pub ` is allowed), spans to the matching closing brace, every node and edge EXACT or (unclassifiable) OPAQUE + UNKNOWN, CALLS / USES / CONSTRAINS edges with the line of the use, refusal past 1024 nodes or 4096 edges. On the tutorial 08 snapshot it equals the hand-read fixture on all 14 declaration nodes and all 9 EXACT edges; it does not emit the fixture's `std.testing` UNKNOWN node, nor its 7 DERIVED edges (4 data flow, 3 tested_by), because a scanner never derives; the derive pass below adds them. Limits: textual matching without scope or type resolution, no block comments, no struct/enum bodies, no cross-spec `use` resolution, capacity 1024 nodes and 4096 edges, u16 indices (all 2225 specs in specs/ fit, largest 775 nodes and 3194 edges after derive; anything past the cap is refused, never truncated), takes a `[]const u8` slice rather than `[65536]u8` plus a length. AST/IR audit is trinity#1650, pilot choice #1651 |
 | Graph IR derive pass | `specs/t27f/spec_graph_derive.t27` (t27#10191) | spec with tests: `derive_graph(g, src)` appends what the exporter must not claim and changes nothing it holds. R1 DERIVED `tested_by` = every EXACT CALLS edge from a test, invariant or bench to a fn read backwards (no line). R2 DERIVED `data_flow` = a const or `use` that an EXACT USES edge ties to a fn and that is named inside one of the fn's return statements (from `return` to the next `;`), on the line of the first such name. R3 UNKNOWN opaque node = an undeclared qualified call target of three or more segments (`std.testing.expectEqual(` names `std.testing`), spanning the lines that call through it. Exporter plus derive equal the tutorial 08 fixture as sets: 15 of 15 nodes, 16 of 16 edges, no special case. Limits: names matched as text; call results and flow through locals are not derived; R3 needs 3 or more segments (two cannot be told from a method call); a graph that does not validate, or a source of another line count, comes back unchanged; past 1024 nodes or 4096 edges the result is refused with the too-many verdict. Not wired to a host: the MCP tool does not call it |
 | UI plan IR + trusted widget catalog | `specs/t27f/ui_plan.t27` (t27#9962) | spec with tests; no renderer draws a plan yet |
@@ -325,6 +325,30 @@ Before you change a uniqueness rule, measure what it can draw from:
   shell heredoc, such as a PR body that mentions tests. Write the body with the Write tool, then
   pass `--body-file`.
 
+## Check the whole corpus for import cycles (t27#10693)
+
+dependency-cruiser's `no-circular` and madge's `--circular` list every module on a cycle.
+`deps_cycles` in `specs/t27f/spec_deps.t27` gives the same answer for the `use-edges` list.
+No t27c subcommand runs a main (t27#10692), so the only host today is a scratch test on the
+Railway lab. Do not commit it.
+
+1. On the lab, write `t27c use-edges --specs-dir specs` to a file **before** you put the
+   scratch spec in `specs/`. Otherwise its own `use` lines join the list, which is harmless but
+   changes the count.
+2. Build the scratch spec from three parts:
+   - a head: `use policy::text_lines; use ci::affected; use claw::t27f_gate; use
+     t27f::spec_deps;`, then a test that opens `const g : str = "`;
+   - the edges, converted with `sed 's/\t/\\t/g; s/$/\\n/' | tr -d '\n'` (in an `fs-lab.sh`
+     string the backslashes are doubled again);
+   - a tail that closes with `$";` and asserts on the result.
+   Room: the 2462-line corpus fits `set` 16384, `done` 65536 and `out` 4096.
+3. Assert `w <= cap` first, so that a refusal is told apart from a wrong list. Then run a
+   **negative control**: the same edges with the opposite expectation must FAIL. On master
+   f7efbc232 the list was exactly `backend.t27` and `rtl.t27`. With #10689 it was empty, and
+   the control that expected the pair failed.
+4. Remove the scratch spec, and `git checkout` any spec you swapped in, before you seal
+   anything.
+
 ## What this skill does not claim
 
 The graph IR is a schema (`specs/t27f/spec_graph.t27`) and `specs/t27f/spec_graph_export.t27` turns source into it, but only exact declarations, spans and edges within the IR's capacity, with `specs/t27f/spec_graph_derive.t27` adding the derived and unknown part by stated rules, and no host calls either yet; no renderer draws a T27F view in the Trinity UI, no
diff --git a/.trinity/seals/t27f_t27f_spec_deps.json b/.trinity/seals/t27f_t27f_spec_deps.json
index 0b91f608f1..a711a4adc5 100644
--- a/.trinity/seals/t27f_t27f_spec_deps.json
+++ b/.trinity/seals/t27f_t27f_spec_deps.json
@@ -15,23 +15,23 @@
     }
   ],
   "config": "gen=c,rust,verilog,zig;tests=zig",
-  "gen_hash_c": "sha256:8dfb9c21bc40906375fafd873fc6a53cbc80726a04beaf7ef3a7248ef6e92b5e",
-  "gen_hash_rust": "sha256:c80e910314b1eb91dff11da6a8b637bcfd411926c341e48b556caa26b394022e",
-  "gen_hash_verilog": "sha256:80dc78cf2dfdbf055cb38f149f2986575b0453994e2db9f9defdb288b3287be4",
-  "gen_hash_zig": "sha256:2a4a4580c07b6aee3d5d235f18adac2eb3ae03a04e03a38d59c25c9184e299ba",
+  "gen_hash_c": "sha256:24377e650e9303759aa6b0b805604ae6d9529c45c10e33242c8be73da902067e",
+  "gen_hash_rust": "sha256:1753610b6d2287d2dec15cd904813eeb22db4208ecfb98b90cec0e34b76274e9",
+  "gen_hash_verilog": "sha256:ca51cd614290ebde62e75fa09618241eb221062882dde6d86248a1c28666469b",
+  "gen_hash_zig": "sha256:add0c9a74743b291f6e2ff110a93868e2a777a7ba99d3126745dc07922ac444e",
   "module": "t27f_spec_deps",
   "ring": 12,
   "seal_schema": 2,
-  "sealed_at": "2026-10-18T22:12:14Z",
+  "sealed_at": "2026-10-18T22:30:43Z",
   "sealed_by": "t27c-bootstrap@0.5.2",
-  "spec_hash": "sha256:75aafa2bc9e6f0c54b721e34d216a498814a097585ba353f72a6aeb813fa4014",
+  "spec_hash": "sha256:b4a87ecfd782bbd552d561fbb39a43b23d8ab4f174b30e1045d52c5cef16d2bb",
   "spec_path": "specs/t27f/spec_deps.t27",
   "tests": {
     "failed": 0,
     "failing": [],
     "forced": false,
-    "passed": 13,
-    "total": 13,
+    "passed": 19,
+    "total": 19,
     "vacuous": 0
   },
   "toolchain": {
diff --git a/specs/t27f/spec_deps.t27 b/specs/t27f/spec_deps.t27
index c2234c881d..e9323116b3 100644
--- a/specs/t27f/spec_deps.t27
+++ b/specs/t27f/spec_deps.t27
@@ -15,15 +15,17 @@
 ;     away. Each spec is in one layer per side, the layer of its shortest path, in edge-list order;
 ;   - every edge drawn is a line of the edge list between two drawn specs, drawn once however often
 ;     the list repeats it, and labelled exact: t27c resolved it, nothing here is inferred;
-;   - the focus is flagged as on a cycle when it imports itself or a spec it imports imports it;
+;   - the focus is flagged as on a cycle when it imports itself or a spec it imports imports it.
+;     deps_cycles asks the same question of every importer and lists each spec flagged, once;
 ;   - a neighbourhood that does not fit the caller's room, or is deeper than the depth asked for, is
 ;     refused whole. Nothing is cut off.
 ; --
 ; The honest limits:
 ;   - a `use` line t27c does not resolve is not in the edge list, so it is not drawn (#7176 counted
 ;     119 such lines in 70 specs);
-;   - the flag says only whether the focus is on a cycle. A cycle among other specs is not reported,
-;     and a spec on a cycle through the focus is drawn on both sides;
+;   - the view flags only the focus. deps_cycles (#10693) lists every spec on a cycle, but nothing runs
+;     it on every PR yet: no t27c subcommand runs a main (#10692). A spec on a cycle through the focus
+;     is drawn on both sides;
 ;   - x and y are numbers for a host, in the spacing of spec_graph_view.t27. No host draws them yet.
 ; --
 ; Measured on the Railway lab, master e607228f7 plus this file: 2461 edge lines over 806 specs. Every
@@ -32,6 +34,9 @@
 ; specs/base/types.t27, with 400 importers. Two specs are flagged on a cycle, and both flags are
 ; true: specs/igla/race/backend.t27 and specs/igla/race/rtl.t27 import each other. `tsort` reports
 ; the same loop on stderr.
+; Measured again for deps_cycles, master f7efbc232 (2462 edge lines), with the list embedded in a
+; scratch test that was not committed: it lists exactly those two specs. With #10689's rtl.t27,
+; which drops its `use igla::race::backend;`, it lists none.
 ; phi^2 + 1/phi^2 = 3 | TRINITY
 
 module t27f_spec_deps;
@@ -154,6 +159,33 @@ pub fn deps_on_cycle(buf: []const u8, n: usize, focus: []const u8, ff: usize, ft
     return false;
 }
 
+/// Every spec of the edge list buf[0..n] that is on a cycle, one per line, into out[0..cap], in the
+/// order of its first line as an importer. A spec that imports nothing is on no cycle, so only
+/// importers are walked, each once (done[0..dcap] holds the ones walked). Returns the bytes written,
+/// or cap + 1 when the list, a walk's set or the done list does not fit, or a walk is deeper than
+/// DEPS_MAX_DEPTH: refused whole, nothing is cut off.
+pub fn deps_cycles(buf: []const u8, n: usize, set: []u8, scap: usize, bounds: []usize, done: []u8, dcap: usize, out: []u8, cap: usize) -> usize {
+    var w : usize = 0;
+    var dn : usize = 0;
+    var s : usize = 0;
+    while (s < n) {
+        const e : usize = line_end(buf, s, n);
+        const t : usize = text_end(buf, s, e);
+        const p : usize = first_tab(buf, s, t);
+        if (p < t and p > s and in_set(done, dn, buf, s, p) == false) {
+            dn = set_add(done, dn, dcap, buf, s, p);
+            if (dn > dcap) { return cap + 1; }
+            if (deps_walk(buf, n, buf, s, p, DEPS_IMPORTS, DEPS_MAX_DEPTH, set, scap, bounds) != DEPS_OK) { return cap + 1; }
+            if (deps_on_cycle(buf, n, buf, s, p, set, bounds[deps_depth(bounds, DEPS_MAX_DEPTH) + 1])) {
+                w = set_add(out, w, cap, buf, s, p);
+                if (w > cap) { return cap + 1; }
+            }
+        }
+        s = e + 1;
+    }
+    return w;
+}
+
 /// Write buf[f..t].
 pub fn deps_put_span(out: []u8, w: usize, cap: usize, buf: []const u8, f: usize, t: usize) -> usize {
     var p : usize = w;
@@ -439,3 +471,73 @@ test negative_control_a_view_that_does_not_fit_returns_more_than_cap {
     assert deps_view(g, lit_len(g), "a", 0, 1, 8, &imp, 64, &ib, &rev, 64, &rb, &es, 128, &out, 32) > 32;
     assert deps_view(g, lit_len(g), "a", 0, 1, 8, &imp, 64, &ib, &rev, 64, &rb, &es, 4, &out, 32) > 32;
 }
+
+// The corpus question #10682 asked one spec at a time: which specs are on a cycle?
+test the_old_igla_pair_lists_both_specs_and_not_their_importer {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = "a\tt\na\tb\nb\tt\nb\ta\nc\tb\n$";
+    const w : usize = deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 64);
+    const want : str = "a\nb\n$";
+    assert w == lit_len(want);
+    assert same(&out, 0, w, want, 0, lit_len(want)) == true;
+}
+
+test negative_control_a_dag_lists_no_spec {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = DEPS_FIXTURE;
+    assert deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 64) == 0;
+}
+
+test a_cycle_the_first_spec_is_not_on_lists_only_the_cycle {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = "a\tb\nb\tc\nc\tb\n$";
+    const w : usize = deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 64);
+    const want : str = "b\nc\n$";
+    assert w == lit_len(want);
+    assert same(&out, 0, w, want, 0, lit_len(want)) == true;
+}
+
+test a_self_import_is_a_cycle_of_one {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = "a\ta\n$";
+    const w : usize = deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 64);
+    const want : str = "a\n$";
+    assert w == lit_len(want);
+    assert same(&out, 0, w, want, 0, lit_len(want)) == true;
+}
+
+test two_cycles_and_a_repeated_line_list_each_spec_once {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = "p\tq\nq\tp\nx\ty\ny\tz\nz\tx\np\tq\n$";
+    const w : usize = deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 64);
+    const want : str = "p\nq\nx\ny\nz\n$";
+    assert w == lit_len(want);
+    assert same(&out, 0, w, want, 0, lit_len(want)) == true;
+}
+
+test negative_control_a_list_that_does_not_fit_is_refused_whole {
+    var set : [64]u8 = undefined;
+    var b : [35]usize = undefined;
+    var done : [64]u8 = undefined;
+    var out : [64]u8 = undefined;
+    const g : str = "p\tq\nq\tp\nx\ty\ny\tz\nz\tx\n$";
+    assert deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 9) == 10;
+    assert deps_cycles(g, lit_len(g), &set, 64, &b, &done, 64, &out, 10) == 10;
+    assert deps_cycles(g, lit_len(g), &set, 64, &b, &done, 9, &out, 64) == 65;
+    assert deps_cycles(g, lit_len(g), &set, 3, &b, &done, 64, &out, 64) == 65;
+}