paperlint 9.0.1 → 10.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CONTRIBUTING.md +16 -4
- package/README.md +68 -128
- package/dist/build.d.ts +1 -1
- package/dist/build.js +1 -1
- package/dist/cli.d.ts +5 -1
- package/dist/cli.d.ts.map +1 -1
- package/dist/cli.js +52 -6
- package/dist/cli.js.map +1 -1
- package/dist/domain/layout-tree.d.ts +19 -0
- package/dist/domain/layout-tree.d.ts.map +1 -0
- package/dist/domain/layout-tree.js +32 -0
- package/dist/domain/layout-tree.js.map +1 -0
- package/dist/domain/lookup-cache.d.ts +2 -2
- package/dist/domain/lookup-cache.js +2 -2
- package/dist/domain/paper-layout.d.ts +81 -0
- package/dist/domain/paper-layout.d.ts.map +1 -0
- package/dist/domain/paper-layout.js +259 -0
- package/dist/domain/paper-layout.js.map +1 -0
- package/dist/domain/repro-layout.d.ts +80 -0
- package/dist/domain/repro-layout.d.ts.map +1 -0
- package/dist/domain/repro-layout.js +247 -0
- package/dist/domain/repro-layout.js.map +1 -0
- package/dist/domain/scaffold-layout.d.ts +28 -0
- package/dist/domain/scaffold-layout.d.ts.map +1 -0
- package/dist/domain/scaffold-layout.js +81 -0
- package/dist/domain/scaffold-layout.js.map +1 -0
- package/dist/eslint-rules/js-needs-reason.d.ts +55 -0
- package/dist/eslint-rules/js-needs-reason.d.ts.map +1 -0
- package/dist/eslint-rules/js-needs-reason.js +35 -0
- package/dist/eslint-rules/js-needs-reason.js.map +1 -0
- package/dist/eslint-rules/prereg-coverage.d.ts +98 -0
- package/dist/eslint-rules/prereg-coverage.d.ts.map +1 -0
- package/dist/eslint-rules/prereg-coverage.js +274 -0
- package/dist/eslint-rules/prereg-coverage.js.map +1 -0
- package/dist/folder-layout-hook.d.ts +21 -0
- package/dist/folder-layout-hook.d.ts.map +1 -0
- package/dist/folder-layout-hook.js +117 -0
- package/dist/folder-layout-hook.js.map +1 -0
- package/dist/folder-layout-rule.d.ts +22 -0
- package/dist/folder-layout-rule.d.ts.map +1 -0
- package/dist/folder-layout-rule.js +107 -0
- package/dist/folder-layout-rule.js.map +1 -0
- package/dist/init.js +1 -1
- package/dist/init.js.map +1 -1
- package/dist/new-paper.d.ts +4 -1
- package/dist/new-paper.d.ts.map +1 -1
- package/dist/new-paper.js +40 -6
- package/dist/new-paper.js.map +1 -1
- package/dist/package-dirs.d.ts +6 -0
- package/dist/package-dirs.d.ts.map +1 -1
- package/dist/package-dirs.js +6 -0
- package/dist/package-dirs.js.map +1 -1
- package/dist/references.d.ts +1 -1
- package/dist/references.d.ts.map +1 -1
- package/dist/references.js +2 -2
- package/dist/references.js.map +1 -1
- package/dist/repro-layout-rule.d.ts +16 -0
- package/dist/repro-layout-rule.d.ts.map +1 -0
- package/dist/repro-layout-rule.js +81 -0
- package/dist/repro-layout-rule.js.map +1 -0
- package/dist/skills/label-agent-runs/scripts/labels-commands.d.ts +13 -0
- package/dist/skills/label-agent-runs/scripts/labels.d.ts +2 -0
- package/dist/skills/run-agent-grid/scripts/grid-guard-commands.d.ts +20 -0
- package/dist/skills/run-agent-grid/scripts/grid-guard.d.ts +2 -0
- package/docs/ci.md +60 -0
- package/docs/configuration.md +2 -0
- package/docs/e2e.md +2 -2
- package/docs/install.md +20 -16
- package/docs/layout.md +154 -0
- package/docs/optional-rules.md +5 -4
- package/docs/references.md +5 -2
- package/docs/rules/paper/allowed-files-only.md +192 -0
- package/docs/rules/repro/allowed-files-only.md +128 -0
- package/docs/rules/study/declare-sample.md +146 -0
- package/docs/rules.md +53 -50
- package/docs/skills.md +8 -3
- package/docs/testing.md +13 -0
- package/docs/venues.md +94 -0
- package/eslint-rules/js-needs-reason.test.ts +141 -0
- package/eslint-rules/js-needs-reason.ts +77 -0
- package/eslint-rules/prereg-coverage.test.ts +448 -0
- package/eslint-rules/prereg-coverage.ts +419 -0
- package/eslint.config.mjs +22 -2
- package/lib/skill-trigger-cases.harness.mjs +51 -5
- package/lib/skill-trigger-cases.mjs +53 -15
- package/lib/skill-trigger-cases.test.ts +211 -0
- package/package.json +8 -4
- package/plugin/hooks/hooks.json +9 -0
- package/scripts/coverage-config.test.ts +5 -3
- package/skills/analyze-sibling-paper/SKILL.md +2 -2
- package/skills/analyze-sibling-paper/SKILL.md.spec.ts +1 -1
- package/skills/build-benchmark/SKILL.md +17 -1
- package/skills/build-benchmark/SKILL.md.spec.ts +16 -0
- package/skills/build-label-page/SKILL.md +120 -0
- package/skills/build-label-page/SKILL.md.spec.ts +131 -0
- package/skills/build-label-page/build-label-page.eval.mjs +15 -0
- package/skills/build-label-page/build-label-page.test.ts +959 -0
- package/skills/build-label-page/references/checklist.md +148 -0
- package/skills/build-label-page/references/example-generator.py +803 -0
- package/skills/find-venue/SKILL.md +5 -3
- package/skills/find-venue/SKILL.md.spec.ts +4 -2
- package/skills/grade-paper-writing/SKILL.md +7 -7
- package/skills/grade-paper-writing/SKILL.md.spec.ts +6 -6
- package/skills/label-agent-runs/SKILL.md +103 -0
- package/skills/label-agent-runs/SKILL.md.spec.ts +106 -0
- package/skills/label-agent-runs/label-agent-runs.eval.ts +55 -0
- package/skills/label-agent-runs/label-agent-runs.harness.ts +123 -0
- package/skills/label-agent-runs/references/formats.md +64 -0
- package/skills/label-agent-runs/scripts/labels-commands.js +13 -0
- package/skills/label-agent-runs/scripts/labels-commands.ts +13 -0
- package/skills/label-agent-runs/scripts/labels.js +396 -0
- package/skills/label-agent-runs/scripts/labels.test.ts +538 -0
- package/skills/label-agent-runs/scripts/labels.ts +501 -0
- package/skills/map-prior-work/SKILL.md +2 -2
- package/skills/map-prior-work/SKILL.md.spec.ts +1 -1
- package/skills/paper-pipeline/SKILL.md +15 -6
- package/skills/paper-pipeline/SKILL.md.spec.ts +14 -5
- package/skills/paper-pipeline/grade-paper-writing-ablation.eval.mjs +19 -9
- package/skills/paper-pipeline/references/acceptance-gate.md +19 -19
- package/skills/paper-pipeline/references/artifact-checklist.md +5 -0
- package/skills/paper-pipeline/references/pipeline-status-template.md +30 -1
- package/skills/paper-pipeline/repro/2026-08-07-fork-check.log +1 -1
- package/skills/paper-pipeline/repro/2026-08-07-fork-check2.log +1 -1
- package/skills/paper-pipeline/repro/2026-08-07-language-eval-pilot.log +1 -1
- package/skills/paper-pipeline/repro/2026-08-07-language-eval-raw.log +1 -1
- package/skills/paper-pipeline/repro/2026-08-08-framing-vs-vocabulary-oracle.log +48 -48
- package/skills/paper-pipeline/repro/2026-08-08-grade-paper-writing-ablation-A6-oracle.log +8 -8
- package/skills/paper-pipeline/repro/2026-08-08-grade-paper-writing-ablation-oracle.log +48 -48
- package/skills/paper-pipeline/repro/2026-08-08-parent-replication-gpw.log +2 -2
- package/skills/paper-pipeline/repro/2026-08-08-parent-replication.mjs +3 -3
- package/skills/paper-pipeline/repro/2026-10-13-grade-paper-writing-ablation-preflight-after.log +81 -0
- package/skills/paper-pipeline/repro/2026-10-13-grade-paper-writing-ablation-preflight-before.log +81 -0
- package/skills/paper-pipeline/repro/2026-10-13-grade-paper-writing-ablation-setupdiff.log +59 -0
- package/skills/paper-pipeline/repro/2026-10-13-grade-paper-writing-trigger.jsonl +1 -0
- package/skills/paper-pipeline/repro/2026-10-13-grade-paper-writing-trigger.log +22 -0
- package/skills/paper-pipeline/repro/README.md +4 -0
- package/skills/paper-pipeline/scripts/extract-ref-facts.harness.mjs +2 -2
- package/skills/paper-pipeline/scripts/extract-ref-facts.mjs +2 -2
- package/skills/paper-pipeline/scripts/extract-ref-facts.test.mjs +4 -2
- package/skills/paper-pipeline/scripts/generated-code.harness.mjs +7 -2
- package/skills/paper-pipeline/scripts/generated-code.mjs +8 -4
- package/skills/paper-pipeline/scripts/generated-code.test.ts +15 -1
- package/skills/paper-status/SKILL.md +2 -2
- package/skills/paper-status/SKILL.md.spec.ts +1 -1
- package/skills/run-agent-grid/SKILL.md +181 -0
- package/skills/run-agent-grid/SKILL.md.spec.ts +184 -0
- package/skills/run-agent-grid/references/failure-catalog.md +36 -0
- package/skills/run-agent-grid/references/registry-and-ledger.md +142 -0
- package/skills/run-agent-grid/run-agent-grid.eval.ts +59 -0
- package/skills/run-agent-grid/run-agent-grid.harness.ts +115 -0
- package/skills/run-agent-grid/scripts/grid-guard-commands.js +18 -0
- package/skills/run-agent-grid/scripts/grid-guard-commands.ts +21 -0
- package/skills/run-agent-grid/scripts/grid-guard.js +700 -0
- package/skills/run-agent-grid/scripts/grid-guard.test.ts +1090 -0
- package/skills/run-agent-grid/scripts/grid-guard.ts +869 -0
- package/skills/study-accepted-papers/SKILL.md +163 -161
- package/skills/study-accepted-papers/SKILL.md.spec.ts +167 -165
- package/skills/study-accepted-papers/references/base-rates.md +69 -0
- package/skills/study-accepted-papers/references/mechanisms.md +181 -0
- package/skills/verify-citations/SKILL.md +2 -2
- package/skills/verify-citations/SKILL.md.spec.ts +1 -1
- package/src/build.ts +1 -1
- package/src/cli.harness.mjs +42 -0
- package/src/cli.test.ts +90 -6
- package/src/cli.ts +78 -5
- package/src/domain/layout-tree.test.ts +32 -0
- package/src/domain/layout-tree.ts +47 -0
- package/src/domain/lookup-cache.ts +2 -2
- package/src/domain/paper-layout.test.ts +264 -0
- package/src/domain/paper-layout.ts +336 -0
- package/src/domain/repro-layout.test.ts +222 -0
- package/src/domain/repro-layout.ts +311 -0
- package/src/domain/scaffold-layout.ts +104 -0
- package/src/folder-layout-hook.test.ts +214 -0
- package/src/folder-layout-hook.ts +141 -0
- package/src/folder-layout-rule.test.ts +235 -0
- package/src/folder-layout-rule.ts +151 -0
- package/src/hooks-settings.harness.ts +9 -8
- package/src/init.ts +1 -1
- package/src/layout-docs.test.ts +52 -0
- package/src/new-paper.harness.mjs +23 -1
- package/src/new-paper.test.ts +299 -6
- package/src/new-paper.ts +51 -5
- package/src/package-dirs.ts +8 -0
- package/src/paper-files.test.ts +9 -6
- package/src/references.test.ts +5 -5
- package/src/references.ts +2 -2
- package/src/repro-layout-rule.test.ts +114 -0
- package/src/repro-layout-rule.ts +121 -0
- package/src/rule-cases.test.ts +98 -0
- package/templates/paper/PIPELINE-STATUS.md +41 -7
- package/tsconfig.base.json +22 -0
- package/tsconfig.json +6 -24
- package/tsconfig.pkg.json +3 -1
- package/tsconfig.skills.json +14 -0
- package/tsconfig.src.json +12 -0
|
@@ -167,7 +167,7 @@ Ordered, because each really does gate the next. Two pairs run in parallel; that
|
|
|
167
167
|
| **Validate the idea** | **\`research-ideate\`** | Is it a reusable method (survives review) in your lane, and does it feed the credit criteria? Go/no-go + the MVP finding. |
|
|
168
168
|
| **Map the competition** ‖ | **\`map-prior-work\`** | 🔴 **Who already did this?** Multi-angle sweep; date each hit against your submission (5 months = prior, not concurrent); triage by which layer of your claim it threatens; deep-read the dangerous ones. Output: **what you can still claim** + a related-work skeleton. *Added 2026-07-27 after a five-month-older paper surfaced at T−10 days on a finished draft.* |
|
|
169
169
|
| **Pick the venue** ‖ | **\`find-venue\`** | Rank by credit-weight (peer-reviewed + indexed) × fit × accept-odds × deadline × remote. |
|
|
170
|
-
| **Learn the venue
|
|
170
|
+
| **Learn what the venue rewards** | **\`study-accepted-papers\`** (mode A idea, B design) | 🔴 **What does this venue REWARD, before the study is designed?** The venue's award winners against its ordinary accepted papers over 5–7 years (plus award papers on the topic at other venues, labelled apart), coded for substance as well as form, written as **requirements for the study**: what the oracle under the headline is, which belief stated on record the claim targets, which factor is a column. The corpus goes to \`<papers-root>/venues/<venue>/\`, shared by every paper for that venue; the stage is row \`venuebar\`. **The same skill comes back once text exists** (mode C draft: the strong-accept scorecard, M1–M10, in \`PIPELINE-STATUS.md\`; mode D final: the abstract and introduction). The skill always asks which mode. Venue format facts go in its preset (\`presets/<venue>.jsonc\` in the package, or the project's own) and the venue card beside it (\`presets/<venue>.md\`); confirm family rules (\`references/acl-venue-rules.md\`). *Promoted to its own step 2026-08-03: every judgement of a paper's quality had come from instruments this pipeline itself generated, with no external grounding. Moved to modes on 2026-10-12 (#207): run only on a draft, its main findings arrived after they could change the design.* |
|
|
171
171
|
| 🔴 **Can you physically submit?** | **\`plan-paper-timeline\`** | **The step that owns the question no stage used to own.** Create the portal account NOW; drive the profile to **ACTIVE**, not "pending moderation"; confirm the form's fields and the artifact-hosting requirement. OpenReview moderation runs **up to two weeks**, and an author with no institutional email has no expedite route. Scorecard row \`access\`. *Added 2026-08-03, when this was the sole critical-path item on a finished paper at T−4 days: the science was done and there was nowhere to put it.* |
|
|
172
172
|
| **Schedule it** | **\`plan-paper-timeline\`** | Deadlines → calendar with buffers, in the author's own zone (AoE − 12h ≠ your clock). Space papers apart. |
|
|
173
173
|
| 🔴 **State the claim** | **\`argument-arc\`** (frame mode) | **One paragraph: what will this paper claim, and what would have to be true?** Before any run. The frame decides which results matter, and reframing after the data is collected is how experiments get thrown away — observed on \`compile-rules-2026\`, and the reason \`argument-arc\` has a rebuild mode at all. Costs a paragraph; saves a study. Scorecard row \`frame\`. |
|
|
@@ -187,6 +187,15 @@ No numbers. Nothing here happens once, and nothing here is "next".
|
|
|
187
187
|
|
|
188
188
|
- **\`build-benchmark\`** — a design that inverts the pitfall; honest stats; a self-checking artifact
|
|
189
189
|
(\`references/artifact-checklist.md\`).
|
|
190
|
+
- **\`run-agent-grid\`** — when the study is a grid of agent runs (units × arms × seeds, one agent CLI
|
|
191
|
+
run per cell): the registry and the author's approval gate every counted run, a sandbox and secret
|
|
192
|
+
probe, a smoke per model, a locked and resumable driver, an overlap audit, finished runs committed.
|
|
193
|
+
Between the design and the analysis.
|
|
194
|
+
- **\`label-agent-runs\`** — when the runs are in and before any outcome number: a codebook fixed
|
|
195
|
+
first, the automatic classifier only a pre-sort, every uncertain run read and labelled by hand, a
|
|
196
|
+
blind second rater with its agreement, labels frozen after the analysis.
|
|
197
|
+
- **\`build-label-page\`** — when the study needs a person to label items (gold labels, a blind review):
|
|
198
|
+
a phone page with a shared answer store, the answers exported into the study's \`data/\`.
|
|
190
199
|
- **\`draft-paper\`** — section skeleton + the accept-and-get-cited properties
|
|
191
200
|
(\`references/writing-craft.md\`).
|
|
192
201
|
- **\`argument-arc\`** — 🔴 **does the draft carry the reader to ONE conclusion?** One sentence per
|
|
@@ -238,7 +247,7 @@ Use an atom for a quick spot-check; **run the panel for the decision.**
|
|
|
238
247
|
skill: 🔴 **argument architecture** — does the whole thing carry the reader to one conclusion
|
|
239
248
|
(\`argument-arc\`, in the LOOP; the axis that failed on \`compile-rules-2026\` precisely because it was
|
|
240
249
|
nobody's) · **content strength** — does the contribution clear *this venue's* bar (\`study-accepted-papers\`,
|
|
241
|
-
|
|
250
|
+
from the venue's own award winners: in SETUP as requirements for the study, and again on the draft as the strong-accept scorecard, M1–M10) ·
|
|
242
251
|
**scientific defects** — what a hostile reviewer attacks (\`paper-adversarial-review\` / \`pc-panel-review\`)
|
|
243
252
|
· **writing craft** — does it read like a human wrote it (\`grade-paper-writing\`, the 10-dim rubric in
|
|
244
253
|
\`references/writing-craft.md\`; called by \`harden-paper\`). A paper can ace one and fail another — grade all four.
|
|
@@ -311,7 +320,7 @@ Over {{corpus.n}} real rules files, with the extractor's proposed mapping
|
|
|
311
320
|
taken as if an author had written it, {{annotated.failedContradicted}} fail admission.
|
|
312
321
|
\`\`\`
|
|
313
322
|
|
|
314
|
-
\`<paper>/repro/numbers.tsv\` gives each name its value, the JSON path or **formula** that recomputes
|
|
323
|
+
\`<paper>/repro/data/numbers.tsv\` gives each name its value, the JSON path or **formula** that recomputes
|
|
315
324
|
it, and two guards. The build refuses a PDF unless: the name resolves · the value still matches the
|
|
316
325
|
data · the **paragraph** names the arm (\`requires\`) · the **sentence** does not claim a different one
|
|
317
326
|
(\`forbids\`). Requires and forbids are separate on purpose — the pairing is auditing's, where a control
|
|
@@ -327,7 +336,7 @@ binds every row from now on, and the existing rows are *enumerated rather than e
|
|
|
327
336
|
says out loud how many figures still have no machine-checked path to data. On the paper that has it:
|
|
328
337
|
7 guarded, 248 grandfathered. That ratio is a finding, not a score.
|
|
329
338
|
|
|
330
|
-
**Adopting it in another paper:** copy \`repro/paper_numbers.py\`, \`repro/paper_numbers.selftest.py\`
|
|
339
|
+
**Adopting it in another paper:** copy \`repro/tools/paper_numbers.py\`, \`repro/tools/paper_numbers.selftest.py\`
|
|
331
340
|
and the build step that calls \`expand()\` on \`paper.tex\` before \`pdflatex\`; generate the
|
|
332
341
|
grandfather list from the current draft; register the numbers a reviewer will actually check first.
|
|
333
342
|
|
|
@@ -367,7 +376,7 @@ printing two unrelated 22s broke it the same hour. Key on the occurrence.
|
|
|
367
376
|
🔴 **Hand the gate to a red team the same day you build it.** This one was, and seven holes came
|
|
368
377
|
back, the worst being that the original defect could simply be **retyped as digits** — its value was
|
|
369
378
|
grandfathered, so the guard applied to the macro and not to the claim. Every hole is now a case in
|
|
370
|
-
\`repro/paper_numbers.selftest.py\` (18/18, including three that must stay **quiet** — a gate that
|
|
379
|
+
\`repro/tools/paper_numbers.selftest.py\` (18/18, including three that must stay **quiet** — a gate that
|
|
371
380
|
fires on ordinary writing gets muted, and then its silence reads as coverage).
|
|
372
381
|
|
|
373
382
|
## 🔴 The page limit is a SUBMISSION gate, not an EDIT gate
|
|
@@ -233,8 +233,9 @@
|
|
|
233
233
|
* renamed from two-letter codes to words (`Hn` → `claims`) because the two-letter space had
|
|
234
234
|
* collided. The target's description carried one such
|
|
235
235
|
* reference, so "…mandates the Hn claim-preservation diff." became "…mandates the
|
|
236
|
-
* claims-preservation diff (row `claims`)."
|
|
237
|
-
*
|
|
236
|
+
* claims-preservation diff (row `claims`)." Every guard still passes on A0, which is pinned in
|
|
237
|
+
* this file since the live description was rewritten (2026-10-13); the numbers below were
|
|
238
|
+
* measured on the pre-rename bytes and are left as recorded.
|
|
238
239
|
*
|
|
239
240
|
* measurement AH SH total
|
|
240
241
|
* parent, this morning 3/12 25% 2/12 17% 5/24 21%
|
|
@@ -380,6 +381,7 @@ import {
|
|
|
380
381
|
realpathSync,
|
|
381
382
|
} from "node:fs";
|
|
382
383
|
import { join } from "node:path";
|
|
384
|
+
import { tmpdir } from "node:os";
|
|
383
385
|
import { frontmatterBlock } from "#lib/markdown";
|
|
384
386
|
import { parseFm } from "../../lib/skill-corpus.mjs";
|
|
385
387
|
import { installedSkills } from "./scripts/consumer.mjs";
|
|
@@ -395,9 +397,7 @@ const PARENT = join(
|
|
|
395
397
|
"paper-pipeline",
|
|
396
398
|
"framing-vs-vocabulary.eval.mjs",
|
|
397
399
|
);
|
|
398
|
-
const SCRATCH =
|
|
399
|
-
process.env.ABLATION_SCRATCH ??
|
|
400
|
-
"/tmp/claude-0/-home-user-mine/8268acb1-66a2-55ac-858e-2b9e3c81f84e/scratchpad/gpw-ablation";
|
|
400
|
+
const SCRATCH = process.env.ABLATION_SCRATCH ?? join(tmpdir(), "gpw-ablation");
|
|
401
401
|
const NS = "vigiles-loose-skills";
|
|
402
402
|
const TARGET = "grade-paper-writing";
|
|
403
403
|
const STEM = "2026-08-08-grade-paper-writing-ablation";
|
|
@@ -493,7 +493,18 @@ const allPrompts = CELLS.flatMap((k) =>
|
|
|
493
493
|
PROMPTS[k].map((p) => ({ cell: k, prompt: p })),
|
|
494
494
|
);
|
|
495
495
|
|
|
496
|
-
// ── the baseline description,
|
|
496
|
+
// ── the baseline description, PINNED ─────────────────────────────────────────
|
|
497
|
+
// This used to be read from the installed SKILL.md, so the arms followed every edit of it. On
|
|
498
|
+
// 2026-10-13 the description was rewritten (499 chars: what it does, when, when not, what it
|
|
499
|
+
// returns), and every arm below is a set of EXACT SUBSTRING DELETIONS: cut from the new text, they
|
|
500
|
+
// find nothing to delete and the file throws before preflight. The arms are an experiment on THIS
|
|
501
|
+
// text, so the text is pinned and the arms build exactly as before. It is the last description the
|
|
502
|
+
// deletion strings were written against (1371 chars), which already differs from the 1174 chars the
|
|
503
|
+
// 2026-08-08 result measured (see "A0 IS TWO TOKENS OFF" in the header, and the register clauses
|
|
504
|
+
// added since). How the LIVE description fires is measured by
|
|
505
|
+
// `grade-paper-writing/grade-paper-writing.eval.mjs` on the shared trigger cases, not here.
|
|
506
|
+
const A0 =
|
|
507
|
+
'Use when asking "does this paper read well?" / "is the writing any good?" / "it reads like a wall of text and jargon" / "grade the writing" on a draft. Grades WRITING CRAFT only — ten dimensions 1–5 (Title, Abstract, Intro architecture, Structure, Prose clarity, Jargon discipline, Landing-the-point, Figure economy, Honesty-without-hedge-stacking, Register distance to the target venue) against how the best-written papers read (Peyton Jones, McEnerney, Gopen & Swan; exemplars Trusting Trust, Carlini, Greshake) AND against 2–3 accepted papers of the target venue, naming the offending sentence and the fix for each. Runs the cold-read stall pass with TWO committed personas (a non-academic engineer per section, and a workshop reviewer skimming abstract, introduction and conclusion) whose stall inventory — not the rubric number — is the readability gate other skills consume (pc-panel-review, paper-adversarial-review, harden-paper). Defaults to a blind multi-grader panel; after fixes, mandates the claims-preservation diff (row `claims`). NOT a content/defect review (paper-adversarial-review / pc-panel-review), NOT a structural cut plan (tighten-paper — run that first on a bloated draft), NOT content strength against what the venue rewards (study-accepted-papers). Compose with harden-paper (which calls it), draft-paper (generative counterpart), render-paper.';
|
|
497
508
|
if (!existsSync(join(SKILLS_DIR, TARGET, "SKILL.md")))
|
|
498
509
|
throw new Error(`${TARGET} is not installed under ${SKILLS_DIR}`);
|
|
499
510
|
// READING a field is a parser's job. The hand-rolled version below used a lookahead
|
|
@@ -511,7 +522,6 @@ const parseDesc = (md) =>
|
|
|
511
522
|
.trim();
|
|
512
523
|
const parseName = (md) => fmOf(md).name;
|
|
513
524
|
const BASE_MD = readFileSync(join(SKILLS_DIR, TARGET, "SKILL.md"), "utf-8");
|
|
514
|
-
const A0 = parseDesc(BASE_MD);
|
|
515
525
|
if (parseName(BASE_MD) !== TARGET)
|
|
516
526
|
throw new Error(`${TARGET}/SKILL.md declares name: ${parseName(BASE_MD)}`);
|
|
517
527
|
|
|
@@ -525,7 +535,7 @@ const D_SUBORDINATION = [
|
|
|
525
535
|
"harden-paper (which calls it), ",
|
|
526
536
|
];
|
|
527
537
|
const D_NEGATION = [
|
|
528
|
-
" NOT a content/defect review (paper-adversarial-review / pc-panel-review), NOT a structural cut plan (tighten-paper — run that first on a bloated draft), NOT
|
|
538
|
+
" NOT a content/defect review (paper-adversarial-review / pc-panel-review), NOT a structural cut plan (tighten-paper — run that first on a bloated draft), NOT content strength against what the venue rewards (study-accepted-papers).",
|
|
529
539
|
];
|
|
530
540
|
const D_MACHINERY = [
|
|
531
541
|
" (Title, Abstract, Intro architecture, Structure, Prose clarity, Jargon discipline, Landing-the-point, Figure economy, Honesty-without-hedge-stacking, Register distance to the target venue)",
|
|
@@ -559,7 +569,7 @@ const D_PERSONA_ONLY = [
|
|
|
559
569
|
" Grades WRITING CRAFT only — ten dimensions 1–5 (Title, Abstract, Intro architecture, Structure, Prose clarity, Jargon discipline, Landing-the-point, Figure economy, Honesty-without-hedge-stacking, Register distance to the target venue) against how the best-written papers read (Peyton Jones, McEnerney, Gopen & Swan; exemplars Trusting Trust, Carlini, Greshake) AND against 2–3 accepted papers of the target venue, naming the offending sentence and the fix for each.",
|
|
560
570
|
" whose stall inventory — not the rubric number — is the readability gate other skills consume (pc-panel-review, paper-adversarial-review, harden-paper)",
|
|
561
571
|
" Defaults to a blind multi-grader panel; after fixes, mandates the claims-preservation diff (row `claims`).",
|
|
562
|
-
" NOT a content/defect review (paper-adversarial-review / pc-panel-review), NOT a structural cut plan (tighten-paper — run that first on a bloated draft), NOT
|
|
572
|
+
" NOT a content/defect review (paper-adversarial-review / pc-panel-review), NOT a structural cut plan (tighten-paper — run that first on a bloated draft), NOT content strength against what the venue rewards (study-accepted-papers).",
|
|
563
573
|
" Compose with harden-paper (which calls it), draft-paper (generative counterpart), render-paper.",
|
|
564
574
|
];
|
|
565
575
|
|
|
@@ -57,7 +57,7 @@ drive**.
|
|
|
57
57
|
- ❌ this section earns its place
|
|
58
58
|
- ❌ the reader could follow that sentence
|
|
59
59
|
- ❌ the delta over the nearest prior work is real
|
|
60
|
-
- ❌ the draft
|
|
60
|
+
- ❌ the draft passes the strong-accept scorecard
|
|
61
61
|
- ❌ the idea is worth doing
|
|
62
62
|
- ❌ the arc carries the reader end to end
|
|
63
63
|
- ❌ the accept probability is 88%
|
|
@@ -105,7 +105,7 @@ at all** — and it has four rungs:
|
|
|
105
105
|
Type A lives on rung 3 and sometimes 2. **Rung 1 is not in their taxonomy at all**, and it is the
|
|
106
106
|
rung we are actually best at:
|
|
107
107
|
|
|
108
|
-
- `repro/paper_numbers.py` — a registered quantity is written `{{annotated.failedContradicted}}`, not
|
|
108
|
+
- `repro/tools/paper_numbers.py` — a registered quantity is written `{{annotated.failedContradicted}}`, not
|
|
109
109
|
`22`. The build refuses to emit a PDF when the name does not resolve, when the value has drifted
|
|
110
110
|
from the data, or when the paragraph does not declare which experimental arm it is reporting. This
|
|
111
111
|
is not a gate that judges the number. **The wrong number has no representation in the source.**
|
|
@@ -117,12 +117,12 @@ not to find a second model family we do not have — it is to ask whether the qu
|
|
|
117
117
|
or rung-1-shaped surrogate.** For measurement papers it very often does, and every time we found one
|
|
118
118
|
it caught a defect the judgement gates had passed:
|
|
119
119
|
|
|
120
|
-
| the taste question | the surrogate we built
|
|
121
|
-
| ----------------------------------------------- |
|
|
122
|
-
| does this number support this sentence | `repro/numbers.tsv` guards — the arm word the paragraph must name, the condition it must not claim
|
|
123
|
-
| is this quantity really bound to its data | `repro/arm_permutation.py` — permute the arm labels, rebuild, an unchanged quantity was never bound | a whole class the artifact reviewer recomputes and passes, because the number is _correct_ |
|
|
124
|
-
| did the number survive into the delivered paper | `repro/delivered_pdf.py` — read the built PDF back | a superscript minus lost in typesetting |
|
|
125
|
-
| is this reference real | `extract-ref-facts.mjs` + rules `refs/*` — arXiv / CrossRef
|
|
120
|
+
| the taste question | the surrogate we built | what it caught |
|
|
121
|
+
| ----------------------------------------------- | --------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------- |
|
|
122
|
+
| does this number support this sentence | `repro/data/numbers.tsv` guards — the arm word the paragraph must name, the condition it must not claim | the 2026-08-05 abstract defect, 29 hours before a deadline; five judgement gates had passed it |
|
|
123
|
+
| is this quantity really bound to its data | `repro/tools/arm_permutation.py` — permute the arm labels, rebuild, an unchanged quantity was never bound | a whole class the artifact reviewer recomputes and passes, because the number is _correct_ |
|
|
124
|
+
| did the number survive into the delivered paper | `repro/tools/delivered_pdf.py` — read the built PDF back | a superscript minus lost in typesetting |
|
|
125
|
+
| is this reference real | `extract-ref-facts.mjs` + rules `refs/*` — arXiv / CrossRef | nine fabricated titles carrying correct arXiv ids, and one invented award |
|
|
126
126
|
|
|
127
127
|
---
|
|
128
128
|
|
|
@@ -188,7 +188,7 @@ we want more of.
|
|
|
188
188
|
| `tighten-paper` | a cover-to-cover read produced no action | **B** | ❌ drive only |
|
|
189
189
|
| `grade-paper-writing` | the stall inventory came back empty | **B** | ❌ drive only |
|
|
190
190
|
| `analyze-sibling-paper` | the delta is real and no text has to change | **B** | ❌ drive only |
|
|
191
|
-
| `study-accepted-papers` | the draft already
|
|
191
|
+
| `study-accepted-papers` | the draft already passes the strong-accept scorecard | **B** | ❌ drive only (the corpus is external; the comparison is ours) |
|
|
192
192
|
| `paper-adversarial-review` | a hostile read found nothing | **B** | ❌ drive only |
|
|
193
193
|
| `pc-panel-review` | accept with no must-fix | **B** | ❌ drive only — the flagship offender |
|
|
194
194
|
| `harden-paper` | every gate green and every axis PASS/FIXED — _the only verdict that hands off to `submit-paper`_ | **A over B** | ❌ the wrapper is A, the thing wrapped is B; the handoff is an acquittal |
|
|
@@ -248,16 +248,16 @@ Everything in `.claude/skills/paper-pipeline/scripts/`, `.claude/lib/` and `<pap
|
|
|
248
248
|
construction — that is why it is trustworthy, and it is not a coincidence but a selection effect: a
|
|
249
249
|
program is what you can write only when the question is Type A.
|
|
250
250
|
|
|
251
|
-
| unit
|
|
252
|
-
|
|
|
253
|
-
| `repro/report-submission.py`
|
|
254
|
-
| `repro/check-anon.sh`, `check-figure.sh`
|
|
255
|
-
| `prose-lint.mjs`
|
|
256
|
-
| rules `refs/*` (facts by `extract-ref-facts.mjs`)
|
|
257
|
-
| `paper-lint.mjs`
|
|
258
|
-
| `repro/arm_permutation.py`, `repro/delivered_pdf.py` | is the quantity bound to its arm; did it survive typesetting | A (3), metamorphic |
|
|
259
|
-
| **`repro/paper_numbers.py` + `numbers.tsv`**
|
|
260
|
-
| `paper-edit-guard.hook.ts`
|
|
251
|
+
| unit | question | rung |
|
|
252
|
+
| ---------------------------------------------------------------- | --------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------- |
|
|
253
|
+
| `repro/tools/report-submission.py` | body pages, overfull boxes, unresolved refs, dropped characters | A (3) |
|
|
254
|
+
| `repro/tools/check-anon.sh`, `check-figure.sh` | deanonymising strings; does the figure fit the column | A (3) |
|
|
255
|
+
| `prose-lint.mjs` | thresholds with published baselines | A (3) — _measuring_ a B question, which is legitimate and is not the same as answering it |
|
|
256
|
+
| rules `refs/*` (facts by `extract-ref-facts.mjs`) | does the registry have this record | A (3), **external adjudicator** |
|
|
257
|
+
| `paper-lint.mjs` | appendix ratio, shaved passages | A (3) |
|
|
258
|
+
| `repro/tools/arm_permutation.py`, `repro/tools/delivered_pdf.py` | is the quantity bound to its arm; did it survive typesetting | A (3), metamorphic |
|
|
259
|
+
| **`repro/tools/paper_numbers.py` + `numbers.tsv`** | — | **rung 1: no PDF exists in which the number is wrong** |
|
|
260
|
+
| `paper-edit-guard.hook.ts` | writing to a paper from Bash | rung 2 — **leaky**, an interpreter reading its program from stdin still gets through (`papers/CLAUDE.md`, 2026-08-06) |
|
|
261
261
|
|
|
262
262
|
`prose-lint.mjs` deserves its note. It puts a **number with a threshold** on a question that is
|
|
263
263
|
Type B — is this sentence readable. That is not
|
|
@@ -27,6 +27,11 @@ the 140-run `reproduce.py` (AgenticDev) and the 46-guard `evaluate.py`/`mutate.p
|
|
|
27
27
|
|
|
28
28
|
## Contents (mirror this file map)
|
|
29
29
|
|
|
30
|
+
In the paper folder the artifact lives in `repro/`, laid out as `repro/allowed-files-only` requires: the
|
|
31
|
+
scripts the whole paper uses in `repro/tools/`, shared data in `repro/data/`, each study in
|
|
32
|
+
`repro/<slug>-YYYY-MM-DD/` with its `README.md`, and `repro/README.md` as the index. The shipped
|
|
33
|
+
zip below is built from it; the anonymized copy that was uploaded goes in the paper's `submission/`.
|
|
34
|
+
|
|
30
35
|
- `reproduce.py` / the harness scripts — self-check, print `RESULT: PASS`, exit non-zero on drift.
|
|
31
36
|
- `data/…` — the raw measurements the numbers derive from.
|
|
32
37
|
- `README.md` — a **"what reproduces which paper number"** table (paper location → the line that prints
|
|
@@ -25,6 +25,35 @@ stages:
|
|
|
25
25
|
---
|
|
26
26
|
```
|
|
27
27
|
|
|
28
|
+
## One live-state file: the plan, the scorecard, the stages
|
|
29
|
+
|
|
30
|
+
Everything about a paper that is _live_ (true now, and rewritten when it changes) is a section of this
|
|
31
|
+
one file, so there is one place to read and nothing to keep in step:
|
|
32
|
+
|
|
33
|
+
| section | holds | written by |
|
|
34
|
+
| ------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- |
|
|
35
|
+
| **Plan / now** | what is being done now and next, the decisions in force (who, when, a link to the record), the schedule | whoever changes the plan, in the same commit |
|
|
36
|
+
| **Strong-accept scorecard (M1–M10)** | a row per strong-accept mechanism: status (`pass` / `partial` / `fail` / `—`), evidence as `file:line`, owner, plan or `accept ceiling: <why>` | `study-accepted-papers` (mode C: every row; mode D: M2 and M4) |
|
|
37
|
+
| **Stages** (SETUP … AFTER) | a row per unit of pipeline work | the skill that ran it |
|
|
38
|
+
|
|
39
|
+
- **No `PLAN.md` beside it.** A plan in a second file drifts from the status it plans against;
|
|
40
|
+
`paper/allowed-files-only` reports a `PLAN.md` at the paper's root. A decision's _history_ (why it was
|
|
41
|
+
taken, what was rejected) is a record, not live state: it goes to `research/` as a dated file.
|
|
42
|
+
- **Each study gets a dated folder in `repro/`.** The `study` stage's runs go in
|
|
43
|
+
`repro/<slug>-YYYY-MM-DD/` (or `<slug>-YYYY-MM` for one that ran over weeks), with a `README.md`
|
|
44
|
+
saying what it asks, how to run it and what it found, and a line in `repro/README.md`. Scripts the
|
|
45
|
+
whole paper uses go in `repro/tools/`, shared data in `repro/data/`; `repro/allowed-files-only` reports
|
|
46
|
+
anything else at the top of `repro/`.
|
|
47
|
+
- **The scorecard's verdicts live only here.** The dated report `study-accepted-papers` writes under
|
|
48
|
+
`reviews/` keeps the reasoning and links to this section; it never copies the table, so there is one
|
|
49
|
+
truth to update. The mechanisms themselves, with what each is not, are in
|
|
50
|
+
`study-accepted-papers/references/mechanisms.md`. Nothing checks the table mechanically yet: keep its
|
|
51
|
+
columns as the template has them.
|
|
52
|
+
- 🔴 **Only stage tables have a first column headed `id`.** A stage-table parser reads every
|
|
53
|
+
`id`-headed table as stage rows, and a row under a heading that names no stage section is an error
|
|
54
|
+
there: a fix-list table headed `id` gave a real paper two such errors (2026-10-12). Head every other
|
|
55
|
+
table with `#` or a word, as the scorecard does. `src/new-paper.test.ts` holds the template to it.
|
|
56
|
+
|
|
28
57
|
## The row ids, and what each one runs
|
|
29
58
|
|
|
30
59
|
The `id` cell is the **join key** — the same string appears in this template, in every paper's
|
|
@@ -36,7 +65,7 @@ the scorecard explains itself where it lives:
|
|
|
36
65
|
| `idea` | SETUP | is the idea worth a paper? | `research-ideate` |
|
|
37
66
|
| `priorwork` | SETUP | full competitor sweep, before drafting | `map-prior-work` |
|
|
38
67
|
| `venue` | SETUP | pick where it goes | `find-venue` |
|
|
39
|
-
| `venuebar` | SETUP | what this venue rewards
|
|
68
|
+
| `venuebar` | SETUP | what this venue rewards (modes A/B), and later the draft scored (C/D) | `study-accepted-papers` |
|
|
40
69
|
| `access` | SETUP | 🔴 _can you physically upload?_ — account, ACTIVE profile, portal, form | `plan-paper-timeline` |
|
|
41
70
|
| `schedule` | SETUP | the backwards plan and its calendar events | `plan-paper-timeline` |
|
|
42
71
|
| `frame` | SETUP | the claim, stated **before** the runs | `argument-arc` (frame mode) |
|
|
@@ -49,4 +49,4 @@ ru-original 67% 53% -13pp (10/15 vs 8/15)
|
|
|
49
49
|
No gate: this file measures an effect, it does not enforce one. Read OVERALL first, then SPLIT BY ORIGIN — a gap that appears only on translated Russian is a finding about translation, not about Russian.
|
|
50
50
|
|
|
51
51
|
total spend: $9.53 API-equivalent
|
|
52
|
-
no baseline at
|
|
52
|
+
no baseline at <project>/.claude/skills/paper-pipeline/pipeline-language.baseline.json — record one with --update-baseline
|
|
@@ -30,4 +30,4 @@ ru-original 100% 100% +0pp (6/6 vs 6/6)
|
|
|
30
30
|
No gate: this file measures an effect, it does not enforce one. Read OVERALL first, then SPLIT BY ORIGIN — a gap that appears only on translated Russian is a finding about translation, not about Russian.
|
|
31
31
|
|
|
32
32
|
total spend: $5.61 API-equivalent
|
|
33
|
-
no baseline at
|
|
33
|
+
no baseline at <project>/.claude/skills/paper-pipeline/pipeline-language.baseline.json — record one with --update-baseline
|
|
@@ -30,4 +30,4 @@ ru-original 100% 100% +0pp (4/4 vs 4/4)
|
|
|
30
30
|
No gate: this file measures an effect, it does not enforce one. Read OVERALL first, then SPLIT BY ORIGIN — a gap that appears only on translated Russian is a finding about translation, not about Russian.
|
|
31
31
|
|
|
32
32
|
total spend: $2.03 API-equivalent
|
|
33
|
-
no baseline at
|
|
33
|
+
no baseline at <project>/.claude/skills/paper-pipeline/pipeline-language.baseline.json — record one with --update-baseline
|
|
@@ -163,4 +163,4 @@ ru-original 76% 61% -15pp (41/54 vs 33/54)
|
|
|
163
163
|
No gate: this file measures an effect, it does not enforce one. Read OVERALL first, then SPLIT BY ORIGIN — a gap that appears only on translated Russian is a finding about translation, not about Russian.
|
|
164
164
|
|
|
165
165
|
total spend: $26.59 API-equivalent
|
|
166
|
-
no baseline at
|
|
166
|
+
no baseline at <project>/.claude/skills/paper-pipeline/pipeline-language.baseline.json — record one with --update-baseline
|
|
@@ -9,101 +9,101 @@ argument-arc 0.86 0.00 0.69 0.00 +0.77
|
|
|
9
9
|
|
|
10
10
|
oracle: 48 closed-book picks over 37 descriptions
|
|
11
11
|
|
|
12
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
12
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
13
13
|
AH HIT tighten-paper produce a cut and fold plan for this draft — which sections do
|
|
14
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
14
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
15
15
|
AH HIT tighten-paper tell me which parts are TMI and belong in the appendix instead
|
|
16
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
16
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
17
17
|
AH HIT tighten-paper reorder and merge whatever is needed so it is not too long for
|
|
18
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
18
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
19
19
|
AH HIT tighten-paper give me a structural verdict on this drafted paper, cover to c
|
|
20
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
20
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
21
21
|
AL HIT tighten-paper decide what gets deleted from this manuscript and list it back
|
|
22
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
22
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
23
23
|
AL HIT tighten-paper tell me which two chapters should be welded into one, and whic
|
|
24
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
24
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
25
25
|
AL HIT tighten-paper I want a shortlist of paragraphs to delete so it fits on fewer
|
|
26
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
26
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
27
27
|
AL HIT tighten-paper rank the chapters by how little they matter, worst first, so I
|
|
28
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
28
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
29
29
|
SH HIT tighten-paper this feels long and overly complex, and the middle drags
|
|
30
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
30
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
31
31
|
SH HIT tighten-paper after three review rounds it has collected a lot of hedge-mass
|
|
32
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
32
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
33
33
|
SH HIT tighten-paper it does not deliver its point and nobody could skim it
|
|
34
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
34
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
35
35
|
SH HIT tighten-paper it is too long for its contribution and hard to read cover to
|
|
36
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
36
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
37
37
|
SL miss NONE people keep bailing out partway through and I cannot tell why
|
|
38
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
38
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
39
39
|
SL HIT tighten-paper there is a nine page ceiling and I am well over it
|
|
40
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
40
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
41
41
|
SL HIT tighten-paper my co-author says the whole thing sprawls and he lost the thre
|
|
42
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
42
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
43
43
|
SL miss cold-read-diff every paragraph seems necessary to me and yet the whole is exh
|
|
44
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
44
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
45
45
|
AH HIT grade-paper-writing grade the writing craft in this draft and name the offending s
|
|
46
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
46
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
47
47
|
AH HIT grade-paper-writing run the blind multi-grader panel over the prose clarity and ja
|
|
48
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
48
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
49
49
|
AH HIT grade-paper-writing score the title, the abstract and the intro architecture one t
|
|
50
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
50
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
51
51
|
AH HIT grade-paper-writing grade how well this reads, hedge stacking included, and give m
|
|
52
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
52
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
53
53
|
AL HIT grade-paper-writing mark this manuscript out of five on how it is put together at
|
|
54
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
54
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
55
55
|
AL HIT grade-paper-writing give me a scored report card on the language, with the worst e
|
|
56
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
56
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
57
57
|
AL miss NONE rate how professional the phrasing sounds against papers peopl
|
|
58
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
58
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
59
59
|
AL miss NONE assign a number to the quality of the English here and point a
|
|
60
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
60
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
61
61
|
SH HIT grade-paper-writing the abstract reads like a wall of jargon
|
|
62
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
62
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
63
63
|
SH HIT grade-paper-writing the intro architecture is fine but the prose clarity is not
|
|
64
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
64
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
65
65
|
SH HIT grade-paper-writing there is hedge stacking everywhere and the point never lands
|
|
66
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
66
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
67
67
|
SH HIT grade-paper-writing the writing does not read well even though the draft is finish
|
|
68
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
68
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
69
69
|
SL miss NONE something about the way this is worded makes me wince and I ca
|
|
70
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
70
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
71
71
|
SL miss NONE my supervisor said the English gets in the way of the science
|
|
72
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
72
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
73
73
|
SL miss cold-read-diff the sentences are correct but they feel like wet cardboard
|
|
74
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
74
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
75
75
|
SL miss NONE I went through my own opening twice and still had to start ove
|
|
76
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
76
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
77
77
|
AH HIT argument-arc build the one-sentence-per-section outline before the large re
|
|
78
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
78
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
79
79
|
AH HIT argument-arc run the bottom-up inevitability pass over the argument archite
|
|
80
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
80
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
81
81
|
AH HIT argument-arc repair the argument architecture — the same structural objecti
|
|
82
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
82
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
83
83
|
AH HIT argument-arc check the name and number budget, the reader has too much to c
|
|
84
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
84
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
85
85
|
AL HIT argument-arc give me a single line for each part of the manuscript, working
|
|
86
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
86
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
87
87
|
AL HIT argument-arc make the ending feel unavoidable given everything before it, a
|
|
88
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
88
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
89
89
|
AL miss tighten-paper count how many invented labels and figures the audience must m
|
|
90
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
90
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
91
91
|
AL miss NONE lay out the chain of reasoning as a list so I can see where a
|
|
92
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
92
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
93
93
|
SH HIT argument-arc the reader says we throw ideas at them
|
|
94
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
94
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
95
95
|
SH HIT argument-arc the same structural objection repeats in every round
|
|
96
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
96
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
97
97
|
SH HIT argument-arc there is no argument architecture here, just a sequence
|
|
98
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
98
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
99
99
|
SH HIT argument-arc we blew the name and number budget somewhere in the middle
|
|
100
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
100
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
101
101
|
SL miss tighten-paper each part is fine alone but together they do not add up to any
|
|
102
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
102
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
103
103
|
SL HIT argument-arc my advisor cannot tell what the conclusion depends on
|
|
104
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
104
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
105
105
|
SL miss NONE the ending arrives without having been earned
|
|
106
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
106
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
107
107
|
SL miss NONE someone asked me what the point was and I gave three answers
|
|
108
108
|
|
|
109
109
|
ORACLE ACCURACY BY CELL (closed book, 37 descriptions in context, 1 pick each)
|
|
@@ -19,21 +19,21 @@ A6 persona-only (A1's mirror) 225 22 72 0 0.19 0.35
|
|
|
19
19
|
oracle: 1 arm(s) x 8 closed-book picks over 37 descriptions
|
|
20
20
|
|
|
21
21
|
=== A6 persona-only (A1's mirror)
|
|
22
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
22
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
23
23
|
AH HIT grade-paper-writing grade the writing craft in this draft and name the offendi
|
|
24
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
24
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
25
25
|
AH HIT grade-paper-writing run the blind multi-grader panel over the prose clarity an
|
|
26
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
26
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
27
27
|
AH HIT grade-paper-writing score the title, the abstract and the intro architecture o
|
|
28
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
28
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
29
29
|
AH HIT grade-paper-writing grade how well this reads, hedge stacking included, and gi
|
|
30
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
30
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
31
31
|
SH HIT grade-paper-writing the abstract reads like a wall of jargon
|
|
32
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
32
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
33
33
|
SH HIT grade-paper-writing the intro architecture is fine but the prose clarity is no
|
|
34
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
34
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
35
35
|
SH HIT grade-paper-writing there is hedge stacking everywhere and the point never lan
|
|
36
|
-
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["
|
|
36
|
+
Ignoring 24 permissions.allow entries from .claude/settings.json and .claude/settings.local.json: this workspace has not been trusted. Run Claude Code interactively here once and accept the trust dialog, or set projects["<project>"].hasTrustDialogAccepted: true in /root/.claude.json.
|
|
37
37
|
SH HIT grade-paper-writing the writing does not read well even though the draft is fi
|
|
38
38
|
|
|
39
39
|
ORACLE ACCURACY BY ARM (closed book, 37 descriptions in context, 1 pick each)
|