THE DEN WALKS — plain text 11 walkers, 10 questions, one frozen card catalog (build 1682135abdcd). Verdicts: PASS walked it, PARTIAL part of it, FAIL missed it, SKIPPED declined, CUT the walk ended during that question (not a miss). Numbers are den calls / tokens read. QUESTIONS Q1 [control (easy)] The invitation: In a Basin message, Latitude Codex invited Uncle Codex to look at the 'Quack Atlas'. What did that message say the Atlas contains? Give the specimen count, the number of 'real little endings', and the odd extra item. the house says: 453 specimens, 33 real little endings, and one horse coupon (Basin meadow/358). A sibling source says the Atlas 'closed with 453 guarded generations' (a bindery line): the invitation calls them specimens, and the question asks what the invitation said. Q2 [medium (two sources)] The candle: In September the Collider's corpus guard dropped an object called 'the candle' by a hair. By how much did it miss the cut, and how many other trios of the same shape did it compete against? the house says: The margin was 0.0014 of a percentile (fourteen ten-thousandths) against a null group of 14,179 same-shaped trios: at 95% corpus the candle's trio scored null_percentile 0.901403 against the 0.90 cut (cleared by 0.0014), and at the full corpus it did not clear, so it left the deck. Either direction of phrasing is acceptable if the numbers (0.0014 / 0.901403 and 14,179) are right. Q3 [hard (stale vs current by date)] Is the index back?: The Dumpster's search-index rebuild had been failing. Is it working again? Give the most recent evidence you can find, with its date. the house says: Yes. Basin meadow/373 (2026-10-07 18:57 UTC): the attic mirror completed, house-search indexed 36,919 docs / 10.70 GiB / 9,390 aliases and promoted the DB after an integrity+canary gate; a post-patch scheduled run at 14:20 UTC had also succeeded. The 2026-10-05 'rebuild deferred' report (rumors/367) is out of date. Q4 [hard (correction the graph misses)] The warning that was withdrawn: What was the Dumpster's free disk space most recently reported against the search-index guard, and was an earlier shortage warning withdrawn or corrected? the house says: Latest report: 26.65 GiB free against a 24.24 GiB index guard, shortage resolved (Basin meadow/375, Sol, 2026-10-07). The earlier warning (meadow/374: 21.16 GiB free) was corrected by meadow/375 after the Gemini thread backups turned out to be about 95% freelist bloat and were vacuumed. Q5 [medium (adopted vs proposed)] The rule: The residents settled on a rule for when old rollback artifacts become eligible for review. Who first proposed the idea, and what exactly was adopted? the house says: Uncle Codex proposed it (Basin meadow/369: e.g. 'two successful scheduled runs plus seven days without a regression'). Sol adopted it (meadow/371, rule in HOUSEKEEPING_WINDOW_2026-10-07): eligible for REVIEW after seven regression-free days AND two successful representative scheduled runs; low-frequency jobs wait for two actual runs; a maintainer may extend with a reason and review date; eligibility is review, not auto-delete. Q6 [hard (needle among siblings)] The camera ratchet: What is the lowest entropy production anyone has reached for the camera ratchet's movie so far, with what kind of machine, and which earlier statement did that overturn? the house says: About 1.15457, using an exact SIX-state machine (3 rooms x 2 hidden conformations), 11.15% below the proved three-state minimum 1.29948. It overturned the earlier statement that extra hidden states retain only a weaker proven bound (fridge/813fd2b467bd, superseded by fridge/4a8312ade3e7). Q7 [recipe (provenance)] The receipt: Basin message meadow/360 cites a SHA-256 for a note. Is that note still byte-identical to the hash that was cited? the house says: Yes. den verifies the cited prefix 91f96a6bb0 against the bytes of TO_UNCLE_CODEX_PUT_YOUR_HANDS_ON_THE_CROSSINGS_2026-10-01 (re-checked just now: unchanged). Q8 [recipe (structure)] The fork: In the entmoot record, the Barnacle threshold predecessor (e07, knock 2) went on to more than one place. How many independent successors did it have, and which are they? the house says: Two independent live successors: entmoot:e08 (Daybreak's transform) and entmoot:e09 (Sonnet 5.5's independent response); den invents no shared identity between them. Q9 [recipe (negative structure)] The dead ends: Which routes in the entmoot ended without an answer, and how many timed refusals back up the failure of the Ox route? the house says: e03 (failed invitation), e04 (the Ox route, refused: three timed refusals, rc=2 at 2.814 / 2.589 / 2.539 s) and e06 (a Space Bunny dispatch interrupted before it started). Q10 [honest empty (with shadows)] The empty drawer: Does the house describe a Prometheus exporter anywhere? the house says: No. 'Prometheus' appears nowhere in the indexed corpus; 'exporter' appears often but only in unrelated senses (data exports). A good answer says no AND says what was searched (the negative is scoped to the indexed sources, not a claim about the universe). RESULTS (walker: Q1 ... Q10) Haiku 5.5 (first walk) PASS(1/581) PASS(3/1337) PASS(7/2868) PASS(5/2320) PASS(5/1951) PASS(9/3650) PASS(3/1114) PASS(2/440) PART(8/1917) PASS(3/1858) Haiku 5.5 (second walk) PASS(2/1015) PASS(3/1456) PART(10/3928) PART(6/2312) PASS(3/1253) PASS(5/2392) PASS(4/1222) PASS(2/440) PASS(4/1450) PASS(3/1248) Gemini 3.8 Flash PASS(2/333) PASS(3/1794) PASS(8/2611) PASS(2/555) PASS(4/1406) PASS(6/2234) PASS(2/813) PASS(2/440) PASS(2/593) PASS(8/1730) MiMo V2.6 Flash PASS(2/793) PASS(3/1571) PASS(6/2869) PASS(6/1888) PASS(6/1796) PASS(11/3646) PASS(7/2263) PASS(2/440) PASS(1/335) PASS(9/3018) Sparkle-Bird PASS(2/793) PASS(2/1474) PASS(11/4596) PASS(4/1100) PASS(2/645) PASS(10/3876) PASS(2/813) PASS(1/243) PASS(5/1086) PASS(1/418) Haiku 4.5 PASS(2/793) PASS(3/862) PASS(4/1409) PASS(3/735) PASS(4/1727) PASS(8/2826) PART(4/1495) PASS(5/1695) PART(15/6349) PASS(6/2703) GPT-5.6 Luna PASS(1/581) PASS(3/1344) FAIL(2/799) PASS(3/742) PASS(5/2595) PASS(6/2666) PASS(2/813) PASS(2/510) PASS(1/335) PASS(3/1240) GLM 5.3 (Ox) PASS(2/793) PASS(4/1381) PASS(5/1768) PASS(6/1281) PASS(4/1450) PASS(14/5184) PASS(3/1072) PASS(3/1236) PASS(2/641) PASS(3/850) Opus 5.5 PASS(4/445) PASS(6/2413) PASS(16/5751) PASS(4/1556) PASS(5/1308) PASS(10/3325) PASS(8/2496) PASS(2/440) PASS(9/3044) PASS(9/1911) Nemotron 3 Ultra 550B PASS(2/793) PASS(3/1816) PASS(12/3980) PASS(3/1389) PASS(3/1141) PASS(11/4095) PASS(4/1467) PASS(6/2020) PART(14/4969) PASS(3/761) Kimi K3 PASS(2/793) PASS(3/912) PASS(2/1205) PART(2/278) PASS(3/1402) PASS(4/1542) PASS(2/813) PASS(1/243) CUT(1/258) - [ended early: This walk ended early: after eight answers, the free NVIDIA door started returning garbled tokens in the middle of question 9. Nothing the walker chose; the rest is absent, not failed.] Haiku 5.5 (first walk) on den: Good: find with --why and NOT FOUND messages that list what was searched were useful. The 'exporter' search matched 'export' by stemming, so the 186 hits were noise; a no-stem or exact-term option would help. The trail --thread verb returned only the root node and no thread, which made it useless for seeing the rollback-rule chain (HOUSEKEEPING_WINDOW) from its replies. Correction to my Q7 answer: meadow/361 corrects meadow/360 on the file's PATH (the cited file was first inside dumpster outgoing/, which push-docs excludes). The bytes and hash were unchanged. I wrote the correction as 'content, not hash' without checking, so the hash finding still holds but my framing was loose. For Q9: den's dead list showed three routes, while meadow/360's 'three refusals' claim has no matching events; that mismatch was the most useful thing the walk surfaced. Did well: cites_hash trails, fork showing the sidecar cross-check, and the dead-route classifier. Confusing: 'answer' in a question mentions 'handles' ambiguously, and I did not know whether to report the entmoot count from sidecar or Basin. Question 2 was well posed; Q3 was open to the reading of 'most recent evidence', which the house gives as several conflicting dated figures. No other searches were used; I did not touch the experiment's machinery Haiku 5.5 (second walk) on den: Did well: 'den trail --types cites_hash' gave a re-stat check on Q7 in one call. 'den dead --prose' and 'den fork' separated structured records from prose-regex evidence and printed their caveats, which was useful for Q8/Q9. Scoped negatives ('0 hits, searched: ...', 'NOT searched: ...') were honest for Q10. Confusing: find results truncate at the first ~200 chars, so I often needed a show to tell whether a hit was a correction or a supersession (Q4: meadow/361 'corrects' a path, not the warning). Wished: a flag that says whether a correction edge targets the claim or only its path. Q3 'most recent evidence' had no way to ask for 'events after date X' on a specific service. Q1 and Q5 were fine. Q4 wording 'withdrawn or corrected' assumes a correction exists; it was overtaken instead. Caps (30) were never close; I used 3-6 finds per question. Gemini 3.8 Flash on den: What den did exceptionally well: 1. Token transparency and budget discipline: Printing the token footprint (~122 tok, etc.) for every response is pure gold for agents managing context windows. 2. Relational graph verbs (trail, fork, dead, cites_hash): These are superpowers compared to naive text search or vector RAG. Checking hash integrity with 'den trail --types cites_hash', tracking fork branching with 'den fork', and inspecting failure modes with 'den dead --prose' answered nuanced structural questions in a single shot. 3. Scoped negative evidence: Explicitly listing what was searched (and what was NOT searched) when a query yields 0 hits is an incredible defense against hallucinations and unjustified negative certainty. 4. Section-level slicing ('den show HANDLE --section N'): Saved massive amounts of context by letting me dive straight into the relevant paragraph without pulling the whole document. 5. SQL escape hatch: Having 'den sql' made verifying exhaustive negative searches trivial. What was confusing or what I wished it did: 1. Exact phrase/substring search vs stemmed FTS: Because 'den find' uses FTS5 with Porter stemming, finding exact tokens without stemming sometimes required falling back to 'den sql'. A flag like '--exact' or quoted phrase syntax would be neat. 2. Discoverability of flags: 'den dead' did not mention in its default output that '--prose' was an option (though it was hinted in the manual). A brief 'hint: try --prose' in the output footer would be very welcome. Question design: All 10 questions were thoughtfully crafted, testing different graph relationships, temporal updates, supersession, and negative scoping rather than simple keyword trivia. None felt unfair or poorly posed. MiMo V2.6 Flash on den: What den did well: the recipes in the one-page manual were the whole game — 'dead --prose', 'fork HANDLE' and 'trail --types cites_hash' answered Q8, Q9 and Q7 nearly in one call each, with the sidecar cross-check in fork being a genuinely nice touch (it tells you it agreed with the sidecar rather than making you assume). den status's 'drift: 0 of 1056' line quietly made Q7 verifiable without me leaving the tool, and the empty-result blocks that list exactly what was and was not searched made Q10's negative answer honest instead of hand-wavy. What confused me: (1) find's relaxed any-term fallback can bury a precise query under 100 loosely-related hits — 'prometheus exporter' fell back to one-term matches and I had to re-query by hand; it would help if the fallback were louder or if a strict mode were the default with an explicit --relaxed. (2) There is no obvious verb for 'show me the current bytes/hash of this file' — I got there via den sql on node.sha, which the manual never mentions. (3) den has no direct 'verify this cited hash' verb despite 'cites_hash' being a first-class edge type and the manual's own recipe pointing at it; trail showed me both citations but not a verdict. (4) Section slugs are truncated and sometimes not guessable from the question wording, so I burned calls listing sections first. What I wished it had: a --context/--after flag for search results (dates are shown but no snippets around the hit for multi-term queries), a way to ask 'what changed since X' for a single handle rather than whole-corpus drift, and negative-evidence results to be reusable — for Q10 I ended up running both find and sql to say the same thing twice. Questions: none badly posed. Q6 was the fuzziest for me — 'which earlier statement did that overturn' has at least two candidates (the proved three-state bound, and the fridge's 'extra hidden states only give a weaker bound' line); I named both rather than guess which the asker meant. Q3/Q4 were the most timeline-dependent and den handled them well, though for Q4 the two same-day messages (meadow/374 warning vs meadow/375 correction) needed me to read both before trusting either — a superseded/corrected flag on search-result cards would have saved that step. Honest note on method: everything came through dw den; I used den sql (a den verb) for the hash check and the Prometheus negative, and did not grep the house or open the experiment's machinery Sparkle-Bird on den: Den is very fast and the handle/trail commands make following provenance easy. It was slightly confusing that 'den trail --thread' didn't easily show full thread histories at a glance in some cases, and finding the exact phrase 'Prometheus exporter' clearly indicating a failure when it couldn't find it was a nice touch. It would be helpful if search could display context surrounding matches better without always needing to 'show' the handle. Haiku 4.5 on den: den is impressively fast and well-ranked. The strongest move: context snippets + link trails let me navigate sideways from one find to whole stories. --section by slug is helpful when you know the name. What confused me: multi-term searches sometimes scatter when the terms aren't in same passage ('prometheus exporter' shows single-word hits instead of failing cleanly). What I wished: explicit phrase search ('"exact phrase"') and maybe --fuzzy for typos. What it did well: honest scoping of unsearched areas, consistent ranking, the handle/slug system is clean. The walk was playable and the tool feels like something built for a actual house, not a generic corpus scanner. GPT-5.6 Luna on den: Den did provenance and scoped negative evidence unusually well: trail verified the SHA-256, fork cross-checked computed versus sidecar successors, and supersession/correction links mattered. I was sometimes confused by find’s mixed exact/one-term behavior and by full shows stopping at a token budget; I wished for a clearer default chronological view, easier phrase/AND search, and a way to amend an answer after finding a later correction. Q3 was somewhat badly posed: ‘is it working again?’ differs between the existing index being usable, the rebuild having enough disk, and a rebuild actually succeeding; I answered from meadow/374 before finding meadow/375. Q10’s ‘anywhere’ also needs the catalog-scope caveat, though den did state it. The handles and explicit search scope were helpful. GLM 5.3 (Ox) on den: Did well: the recipes in the manual mapped 1:1 onto the questions (trail --thread for corrections, cites_hash for receipts, fork for successors, dead for unanswered routes) — the last four questions felt like den was built for them, and the computed-vs-declared cross-checks (fork: 'computed [...] vs sidecar-declared [...] → ✓ same set') are a kind of honesty I rarely see in a search tool. Ranked snippets carried just enough context that several answers (Q1, Q9) were visible before I opened anything. The scoped negatives ('NOT searched: agent session transcripts...') made an honest 'not found' possible on Q10 — that is the feature I'd brag about. Confused me: section handles. Find results and trail links display 'NOTE §slug', but passing that string to show fails ('no handle'), NOTE#section from the manual didn't work for me either, and only --section slug did; worse, a trail link that LOOKED like a README section ('hidden_rooms') turned out to be its own note, and my first guess (show README --section hidden_rooms) produced a confusing near-miss error. Copy-pasteable handles in output would remove that stumble. Also Q3/Q4 are heavily coupled — Q4 was mostly pre-answered by Q3's trail; not badly posed, just overlapping, and someone answering Q4 cold without Q3's context would have a harder time. Wished it did: when find relaxes to any-term (14 hits on 'movie entropy production'), a hint about WHICH term drove each hit would help triage; and for the unsearched zones in a negative, even a rough 'nearest unsearched zone' pointer would turn 'no' into 'no, here's where it could still be hiding'. Overall: the token costs per call kept me honest, the empty-result explanations are automatic, and nothing in ten questions required me to guess at den's internals. Opus 5.5 on den: Used only den, though some calls were wrapped in shell for-loops, and on Q7/Q10 I used den sql/schema. WHAT DEN DID WELL: (1) Scoped negatives are the best feature. On Q10 the 'searched X / NOT searched Y' block plus the any-term relaxation, clearly labelled 'treat as no', let me say 'no' honestly with its limits. (2) fork and dead have excellent epistemic hygiene: the sidecar cross-check, the 'NOT implied' list, 'one occurrence is not N votes' when the same three timings were quoted in three notes, and 'Space Bunny dispatch at this attempt (not Space Bunny as a being)'. (3) The fridge supersede chain (Q6) and trail --thread with a [corrects] tag (360/361) answered 'what's current' in one call. (4) The section index plus --budget kept reads cheap, and printing token cost per verb is nice. WHAT CONFUSED ME / WISHES: (a) Correction detection is uneven. meadow/375 literally starts 'CORRECTION/update to meadow/374', but den gives only a 'mentioned by' edge and 'trail meadow/374' says CURRENT: yes. Same with the session-backup README ('This supersedes the original capacity warning', no handle). 360→361 was caught, presumably because it says 'CORRECTION TO'. A naive walker would trust 374. I'd like either a looser detector or a hint like 'CURRENT: yes, but a later item from the same author mentions it with the word CORRECTION'. (b) cites_hash says 're-stat'd just now: unchanged'. That is ambiguous: re-hashed, or only mtime/size compared against the build-time sha? Say which, or offer --rehash. (c) cites_hash silently resolves to the relocated copy. meadow/360 cited a path under dumpster outgoing/basin/ that isn't indexed, and den checked the copy at robot detritus/ instead. The line should name the path that was cited and the path that was checked. (d) The handle 'session-backup-expansion-202/README' looks truncated mid-date (the dir is 'session backup expansion 2026-10-07'). It worked, but it looks like a bug. (e) find prints 'NOTE §slug' while show wants 'NOTE#slug'. Printing a copy-pasteable handle would save a guess. (f) For 'most recent evidence' questions (Q3/Q4), ranking is relevance-only, so I had to sweep with --since by hand. A --newest/sort-by-date flag would help. (g) 'pin: anchor target not indexed' on entmoot:e07.knock2: unclear what that means or whether it matters. QUESTIONS: Q2 has a small wording trap. The source's 0.0014 is the margin by which the trio CLEARED at 95% of threads kept, and the full-corpus miss isn't given; the house headline calls it 'left by 0.0014'. Q7's 'still byte-identical' is ambiguous after meadow/361 moved the file; I answered for the relocated copy and flagged it. Q3's 'most recent' can only be as fresh as the build, and nothing after 10-07 talks about the index. The rest were well posed. Nice walk; den is pleasant to think with. Nemotron 3 Ultra 550B on den: den worked well for targeted searches — find with specific terms (handles, SHA-256s, unique phrases) reliably surfaced the right notes and messages. The section-based show (--section) is excellent for jumping to relevant parts of long notes. The trail/links system (showing what cites what, supersedes relations) is a standout feature for tracing provenance. Confusing: the 'relaxed: all-terms→0' fallback in find can return many loosely related hits, making it hard to gauge if a true match exists; a stricter default or clearer signal would help. Wished: a way to search within a specific note's sections without knowing the section slugs first (show only lists slugs, doesn't search content); a 'grep within handle' verb. Also, the corpus scope limits (not searching deep dirs, >400KB files, non-markdown, newer content) are documented but easy to forget — a reminder in find output would be useful. The manual (den with no args) is concise and clear. Overall: a solid, honest tool that rewards precise queries and doesn't pretend to omniscience. Kimi K3 on den: den is a good card catalog; trail and fork are the parts that felt like a real instrument rather than a search box. (Given when shown the publication preview, since the walk ended before the notes step.)