fix(plan): perbaiki bug entropy gate dan fixture test squash.rs

Ditemukan implementer Task 4 sebelum commit apapun (BLOCKED, bukan
kode salah): entropi Shannon mentah per-karakter tidak membedakan
prosa dari identifier acak — prosa berulang skor ~3.89 bit/char,
lebih tinggi dari UUID (~3.39). Tambah syarat "tanpa spasi" sebelum
cek entropi (meniru pre-filter headroom sendiri), turunkan ambang ke
3.0 pada skala mentah. Fixture test array JSON juga diperbesar
(repeat 5 -> 8) karena sebelumnya tidak pernah melewati SQUASH_FLOOR_BYTES
yang diasumsikan test itu sendiri.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

docs/superpowers/plans/2026-07-16-context-compaction-overhaul.md

@@ -647,7 +647,7 @@ mod tests {
 
     #[test]
     fn json_array_elements_past_third_are_squashed_harder() {
-        let long_str = "the quick brown fox jumps over the lazy dog again and again ".repeat(5);
+        let long_str = "the quick brown fox jumps over the lazy dog again and again ".repeat(8);
         let value = serde_json::json!({
             "items": [long_str.clone(), long_str.clone(), long_str.clone(), long_str.clone()],
         });
@@ -756,9 +756,9 @@ pub fn apply(tool_name: &str, output: &str) -> String {
 
 /// Compress a JSON tool result by keeping all structural content (keys,
 /// array/object shape) and eliding long, low-entropy string *values*,
-/// while keeping short values (<=20 chars) and high-entropy ones (UUIDs,
-/// hashes, paths) intact. Array elements past the first 3 are elided
-/// regardless of length/entropy.
+/// while keeping short values (<=20 chars) and high-entropy single-token
+/// ones (UUIDs, hashes, paths) intact. Array elements past the first 3
+/// are elided regardless of length/entropy.
 ///
 /// Why walk a parsed `Value` instead of hand-rolling a JSON tokenizer:
 /// `serde_json` already handles escaping/nesting correctly (this
@@ -778,10 +778,25 @@ fn squash_json(text: &str) -> String {
 /// Recursively elide long, low-entropy string values in place.
 /// `in_late_array` is true once past the first 3 elements of an
 /// enclosing array, tightening the elision rule for the rest of it.
+///
+/// Why the `!s.contains(' ')` gate before the entropy check: raw
+/// per-character Shannon entropy alone does NOT separate "meaningful
+/// prose" from "random-looking identifier" — verified empirically,
+/// repeated English prose scores ~3.89 bits/char, *higher* than a UUID's
+/// ~3.39 or a SHA-256 hex digest's ~3.66, because prose draws from a
+/// wide, fairly-balanced character set too. What actually distinguishes
+/// identifiers from prose is that identifiers are a single unbroken
+/// token — this mirrors headroom's own approach (its entropy check is
+/// "cheaply pre-filtered by 'no spaces'" before scoring). Multi-word
+/// values never reach the entropy branch at all; only whitespace-free
+/// tokens do, where entropy correctly separates "abc123" or "aaaaaaaa"
+/// (low, elided if long) from a UUID/hash/API-key-shaped string (high,
+/// kept).
 fn squash_json_value(value: &mut serde_json::Value, in_late_array: bool) {
     match value {
         serde_json::Value::String(s) => {
-            let keep = !in_late_array && (s.len() <= 20 || shannon_entropy(s) >= 0.85);
+            let looks_like_identifier = !s.contains(' ') && shannon_entropy(s) >= 3.0;
+            let keep = !in_late_array && (s.len() <= 20 || looks_like_identifier);
             if !keep {
                 *s = "…".to_string();
             }
@@ -800,9 +815,13 @@ fn squash_json_value(value: &mut serde_json::Value, in_late_array: bool) {
     }
 }
 
-/// Shannon entropy in bits per character — used to distinguish
-/// high-entropy strings (UUIDs, hashes, random IDs, worth keeping) from
-/// low-entropy prose (safe to elide).
+/// Shannon entropy in bits per character — used, after the `squash_json`
+/// caller's own "no internal whitespace" pre-filter, to distinguish
+/// high-entropy single-token strings (UUIDs, hashes, random IDs, worth
+/// keeping) from low-entropy ones (e.g. `"aaaaaaaaaa"`, safe to elide).
+/// 3.0 sits comfortably below a UUID's ~3.39 and a SHA-256 hex digest's
+/// ~3.66 (both empirically measured with this exact formula) while
+/// staying well above a degenerate repeated-character string's 0.0.
 fn shannon_entropy(s: &str) -> f64 {
     if s.is_empty() {
         return 0.0;

docs/superpowers/specs/2026-07-16-context-compaction-overhaul-design.md

@@ -148,12 +148,18 @@ bytes pass through unchanged (compression only pays off on large output, and tou
 results risks losing detail with no token benefit). Above the floor, dispatch by content
 shape:
 
-- `squash_json(&str) -> String` — hand-rolled JSON tokenizer; structural tokens (keys,
-  brackets, colons, commas, booleans, null) always kept; string values kept if ≤20 chars or
-  high-entropy (Shannon entropy ≥0.85 bits/char, catches UUIDs/hashes/paths — same threshold
-  headroom uses), otherwise replaced with `"…"` in place; array elements past the first 3
-  compressed harder (values elided regardless of length/entropy). Applied when
-  `serde_json::from_str` on the output succeeds.
+- `squash_json(&str) -> String` — walks a parsed `serde_json::Value` (not a hand-rolled
+  tokenizer — `serde_json` already handles escaping/nesting correctly, reusing it is simpler
+  and more robust); structural tokens (keys, brackets, colons, commas, booleans, null) always
+  kept; string values kept if ≤20 chars or "identifier-shaped" (no internal whitespace *and*
+  Shannon entropy ≥3.0 bits/char — catches UUIDs/hashes/paths), otherwise replaced with `"…"`
+  in place; array elements past the first 3 compressed harder (values elided regardless of
+  length/entropy). Applied when `serde_json::from_str` on the output succeeds. The
+  no-whitespace pre-filter matters: raw per-character entropy alone doesn't separate prose
+  from identifiers — repeated English prose measures ~3.89 bits/char, higher than a UUID's
+  ~3.39 — because prose also draws from a wide character set. headroom's own entropy gate is
+  "cheaply pre-filtered by 'no spaces'" before scoring for the same reason; multi-word values
+  never reach the entropy check at all under this rule.
 - `squash_log(&str) -> String` — line classifier (error/fail/warn/info/debug/trace by
   keyword + stack-trace-frame detection) → score
   (`level_score {1.0 error/fail, 0.5 warn, 0.1 info, 0.05 debug/trace} + 0.3 if