paperlint 7.1.0 → 8.0.0

package/docs/prior-art/paper-folder-scaffolding.md

@@ -16,10 +16,6 @@ still left to a human copying something by hand?**
 > - `--hooks=local` is **deferred, not implemented**; `init` refuses it by name.
 >
 > Where the build differed from this text, it is listed at the end, in § 8.
->
-> **Later, 2026-09-24:** Markdown papers are deprecated and being removed
-> ([#57](https://github.com/zernie/research-paper-pipeline/issues/57)). Where this record mentions
-> `paper.md` or `--format md`, it describes the tool as it was when written.
 
 1. A paper folder needs a `PIPELINE-STATUS.md`, and today nothing makes it — you copy a template.
    Should there be a command, and where should per-paper settings live: in a file in every paper
@@ -66,11 +62,11 @@ changes, no migration.
   (`src/init.ts:448`), offers a CI workflow, reports missing programs, and **prints** the two
   `/plugin` lines (`src/init.ts:257-258`). It never creates the papers directory and never creates
   a paper.
-- **A directory is a paper** if it holds any of `PIPELINE-STATUS.md`, `paper.tex`, `paper.md`,
-  `venue.json` (`src/build.ts:41-46`, `src/structure.ts:39`). Detection is generous.
-- **`rpp lint` requires `PIPELINE-STATUS.md`** in every such directory, plus one of
-  `paper.tex`/`paper.md` (`src/structure.ts:38-43`). A paper with only `paper.md` fails the very
-  first `rpp lint` with `missing PIPELINE-STATUS.md` — that is the README's own "First run" example.
+- **A directory is a paper** if it holds any of `PIPELINE-STATUS.md`, `paper.tex`, `venue.json`
+  (`src/build.ts`, `src/structure.ts`). Detection is generous.
+- **`rpp lint` requires `PIPELINE-STATUS.md`** in every such directory, plus `paper.tex`
+  (`src/structure.ts`). A paper with only `paper.tex` fails the very first `rpp lint` with
+  `missing PIPELINE-STATUS.md` — that is the README's own "First run" example.
 
 ⚠️ **One inconsistency worth knowing.** `paper/stages` says in its header that a missing `stages`
 field is _legitimate_ — "a paper that has shipped nothing owes nothing" (`eslint-rules/paper-stages.mjs:24-27`) —
@@ -101,7 +97,7 @@ human to copy and tick them.
 ## 2. What `PIPELINE-STATUS.md` actually holds — the split
 
 Read from the template (`skills/paper-pipeline/references/pipeline-status-template.md`), the
-fixtures (`fixtures/paper-stages/*`, `fixtures/real-markdown-paper/`) and four real files in the
+fixtures (`fixtures/paper-stages/*`) and four real files in the
 owner's corpus. Real sizes: 86, 95, 1 057 and 1 301 lines. In the two big ones the machine-read
 part is the first 10–25 lines of front matter; the rest is a dated journal of decisions. (No
 content of those files is reproduced here.)
@@ -110,7 +106,6 @@ content of those files is reproduced here.)
 | ------------------------------------------------------------------------------- | --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | paper short name / title                                                        | **config**            | human, once — `rpp new` can prefill from the folder name                                                                                                                  |
 | venue, deadline, blind model (the `Venue:` header line)                         | **config**            | human; changes when the paper is re-targeted                                                                                                                              |
-| format (`.tex` vs `.md`)                                                        | **derived**           | from which source file exists                                                                                                                                             |
 | venue profile, paper kind, pdf path (a separate `venue.json` in one real paper) | **config**            | human; already a per-paper file in practice                                                                                                                               |
 | typography allowance (`typographyDebt`)                                         | **config**            | human, but it sits in the root key keyed by path today                                                                                                                    |
 | `stages[]`: `stage`, `date`, `venue`                                            | **state**             | **tool** at the moment of freezing (`rpp freeze`, a later step)                                                                                                           |
@@ -172,9 +167,9 @@ parsing markdown — the fix is to ship the template as its own file and have th
    (entry front matter), Cargo (member `Cargo.toml`), Eleventy (directory data). The tools that put
    per-unit settings in the root (ESLint `files`, Vale sections) are _linters of code_, where units
    have no identity of their own — a paper does.
-3. **Front matter of the source itself is not available to rpp**, because half the papers are
-   LaTeX, which has no YAML front matter. Quarto and Astro can use the document's own header only
-   because every document is markdown. So rpp needs a per-paper **file**, and it already has one:
+3. **Front matter of the source itself is not available to rpp**, because a paper is LaTeX, which
+   has no YAML front matter. Quarto and Astro can use the document's own header only because their
+   documents carry one. So rpp needs a per-paper **file**, and it already has one:
    `PIPELINE-STATUS.md`, whose front matter is exactly where Quarto would put these fields.
 4. **Root-level keys that are really per-paper** (`typographyDebt: { "papers/my-paper": … }`,
    `src/cli.ts:104`) are the ESLint-overrides style. They work, but they break silently on a folder
@@ -209,9 +204,9 @@ written once at freeze time, then checked forever — and today a human types it
 
 ### Commands
 
-**`rpp new <name> [--format tex|md]`** — creates `<papers>/<name>/` with:
+**`rpp new <name>`** — creates `<papers>/<name>/` with:
 
-- `paper.tex` or `paper.md` (a minimal stub; `--format` defaults to `tex` — ask when interactive);
+- `paper.tex` (a minimal stub);
 - `PIPELINE-STATUS.md` from a **template file** shipped in the package, with the name filled in and
   **no** `stages` field (a new paper has shipped nothing — `paper-stages.mjs:24`).
 
@@ -228,7 +223,7 @@ Rules for it, each taken from a tool above:
 **Later step, NOT part of this change — `rpp freeze <paper> <stage> [--pdf <file>] [--date <iso>]`**,
 the release-please move, recorded so the design is not lost:
 
-- copies the built pdf and the source into `versions/<date>-<stage>.pdf|.tex|.md`;
+- copies the built pdf and the source into `versions/<date>-<stage>.pdf|.tex`;
 - **appends** one record to `stages` with `stage`, `date`, `pdf`, `bytes`, `source`, `sourceBytes`;
 - refuses a stage outside `STAGES` (`paper-stages.mjs:53`) and a date/stage pair already present.
 
@@ -259,15 +254,15 @@ the skills tell agents to update them.
 
 - **Existing folders with a `PIPELINE-STATUS.md`** (the owner's four): nothing changes. Their
   front matter is already the per-paper carrier; a later `rpp freeze` would append to their lists.
-- **Existing folders without one** (a draft that has only `paper.md`): `rpp new <that-name>` fills
-  the missing file and touches nothing else.
+- **Existing folders without one** (a draft that has only `paper.tex`): `rpp new <that-name>`
+  fills the missing file and touches nothing else.
 - **No `rpp migrate` is needed** for this recommendation. One would be needed only if the carrier
   later moves to `paper.yml`.
 
 ### The smallest first step
 
 Move the template out of the code fence into its own file (say
-`templates/paper/PIPELINE-STATUS.md`, plus `paper.tex` and `paper.md` stubs), make
+`templates/paper/PIPELINE-STATUS.md`, plus a `paper.tex` stub), make
 `pipeline-status-template.md` point at it instead of containing it, and add `rpp new` that copies
 it. Then the README sentence "Nothing generates it and no command refreshes it" becomes "`rpp new`
 creates it". No rule changes, no consumer migration. `rpp freeze` is step two; `init` offering to
@@ -435,13 +430,13 @@ unset **and** no `--yes`._
 
 ### What `init` asks, and what it does without asking
 
-| decision          | human at a terminal                                                                                         | agent / CI / `--yes`                                                                              | flag                                    |
-| ----------------- | ----------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- | --------------------------------------- |
-| papers directory  | asked only if several candidates (today)                                                                    | first candidate, said so (today)                                                                  | `init <dir>` then the key               |
-| **hooks**         | "Wire the three paper hooks into `.claude/settings.json` (committed, shared)? **[Y/n]**"                    | **yes** — the guard is what the package is for, the edit is idempotent and shows up in `git diff` | `--no-hooks` (`--hooks=local` deferred) |
-| CI workflow       | "Add a GitHub Actions workflow? **[y/N]**" (today)                                                          | **no** (today) — it writes into `.github/` and spends Actions minutes, which are metered          | `--ci` / `--no-ci`                      |
-| first paper       | only when the papers dir holds no paper: "Create a first paper? name: [skip]" then "format: tex / md [tex]" | **no**, unless `--paper <name>` is given                                                          | `--paper <name> --format tex\|md`       |
-| external programs | reported (today)                                                                                            | reported (today)                                                                                  | —                                       |
+| decision          | human at a terminal                                                                      | agent / CI / `--yes`                                                                              | flag                                    |
+| ----------------- | ---------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- | --------------------------------------- |
+| papers directory  | asked only if several candidates (today)                                                 | first candidate, said so (today)                                                                  | `init <dir>` then the key               |
+| **hooks**         | "Wire the three paper hooks into `.claude/settings.json` (committed, shared)? **[Y/n]**" | **yes** — the guard is what the package is for, the edit is idempotent and shows up in `git diff` | `--no-hooks` (`--hooks=local` deferred) |
+| CI workflow       | "Add a GitHub Actions workflow? **[y/N]**" (today)                                       | **no** (today) — it writes into `.github/` and spends Actions minutes, which are metered          | `--ci` / `--no-ci`                      |
+| first paper       | only when the papers dir holds no paper: "Create a first paper? name: [skip]"            | **no**, unless `--paper <name>` is given                                                          | `--paper <name>`                        |
+| external programs | reported (today)                                                                         | reported (today)                                                                                  | —                                       |
 
 Why the hooks default is "yes" even without asking, while CI is "no": the rule `install.md` already
 adopted from Playwright is _"ask only about what cannot be guessed or is expensive"_. Installing a
@@ -495,8 +490,8 @@ Recorded so the design and the code do not quietly disagree.
 - **`rpp new` lints the new folder itself** (`rpp lint <papers>/<name>`), so the structure check —
   which looks at the subdirectories of the path it is given — does not run there; the folder was
   just made with the required files, so there is nothing for it to find.
-- **Interactive `init` asks "Create a first paper? name: [skip]" and then the format** only when the
-  name was given and `--format` was not.
+- **Interactive `init` asks "Create a first paper? name: [skip]"** only when the papers directory
+  holds no paper.
 
 ## Sources checked
 

package/docs/rules/cycle/evidence.md

@@ -0,0 +1,87 @@
+# cycle/evidence
+
+**Level:** error · **Reads:** the paper's `paperlint.json` (`cycles`) and the paper folder ·
+**Reported on:** the first line of the paper's `PIPELINE-STATUS.md`
+
+## What it catches
+
+A closed cycle — `outcome.kind` is `accepted`, `rejected` or `withdrawn` — names an `evidence`
+file that is not in the paper folder: the decision mail, the reviews, or the note that records the
+withdrawal was never saved, or the path is wrong.
+
+The same for a deadline's `override`: the instant a human decided binds instead of what the pages
+said rests on its `evidence` — the chairs' mail that extended the deadline, a saved page — and that
+file must be in the paper folder.
+
+## Why
+
+The outcome of an attempt is a claim about what a venue decided, and the record is read by people
+and programs that were not there when the mail arrived: a status command, a scorecard check, the
+next attempt's prior-work delta. A decision whose evidence is not in the folder is a claim with
+nothing behind it, and the day it is doubted there is nowhere to look. The record therefore refuses
+an outcome without `evidence` at all, and this rule requires the file to exist.
+
+## Examples
+
+Failing — `paperlint.json`:
+
+```json
+{
+  "cycles": [
+    {
+      "id": "first-2026",
+      "venue": { "kind": "preset", "extends": "paperlint:acm-sigconf" },
+      "opened": "2026-07-01",
+      "outcome": {
+        "kind": "rejected",
+        "date": "2026-09-08",
+        "evidence": "reviews/decision.md"
+      }
+    }
+  ]
+}
+```
+
+with no `reviews/decision.md` in the paper folder:
+
+```text
+papers/my-paper/PIPELINE-STATUS.md
+  1:1  error  cycle «first-2026» is rejected (2026-09-08) with evidence `reviews/decision.md`, which is
+              not in the paper folder — save the decision mail, the reviews or the note that records
+              the decision there, or correct the path  cycle/evidence
+```
+
+An override whose evidence was never saved:
+
+```text
+  1:1  error  cycle «msr-2027»: the override of the "submission" deadline (2026-10-27T04:00:00Z)
+              rests on `mail/extension.eml`, which is not in the paper folder — save the mail or the
+              page that grants it there, or correct the path  cycle/evidence
+```
+
+Passing:
+
+- the same record with `reviews/decision.md` present;
+- an open cycle (`"outcome": { "kind": "open" }`, or no `outcome`): nothing has been decided;
+- a paper whose `paperlint.json` has no `cycles`.
+
+## Options / preset fields
+
+No rule options. `evidence` is a path relative to the paper folder; it may name any file — the
+saved mail, a markdown note with the reviews, a PDF.
+
+## What it does not check
+
+- What the file says. A rejection recorded as `accepted` with the rejection mail as evidence passes
+  here; the person who records the outcome reads the mail. Likewise an override's mail is not read
+  for its date.
+- The `date`. It must be a day (`YYYY-MM-DD`), nothing more.
+- Whether a `desk` rejection is marked as such (`"desk": true`); it changes what work follows, not
+  whether the record is consistent.
+
+## How to fix
+
+Save the decision in the paper folder — the mail as it arrived, the reviews as the portal shows
+them, or a dated note that records a withdrawal and why — and name that file in `evidence`. If the
+file exists under another name, correct the path. For an override, save the mail or page that moved
+the deadline and name it in the override's `evidence`.

package/docs/rules/cycle/port-done.md

@@ -0,0 +1,75 @@
+# cycle/port-done
+
+**Level:** error · **Reads:** `paper.tex` → the paper's `paperlint.json` (`cycles`) → the venue
+preset's `template` · **Reported on:** `paper.tex`
+
+## What it catches
+
+The current cycle declares `"phase": "porting"` — the source is still in the previous venue's
+template, so the venue's format rules are silent — and the paper's `\documentclass` is already the
+one the venue's preset requires. The port is done and the declaration is stale.
+
+## Why
+
+`"phase": "porting"` is a record of open work, not a softer severity: the day a new venue is chosen
+the source is still in the old venue's template, and a page count under the wrong template says
+nothing about the new limit. While the phase is declared, every `pdf/*`, `format/*`, `anonymity/*`
+and `tex/*` venue rule stays silent for that paper. A declaration that silences rules must not be
+able to stay on after its reason is gone: once the class line is the venue's, this rule fires, and
+the only way to be quiet again is to drop the phase — which turns the format rules back on.
+
+## Examples
+
+Failing — `paperlint.json`:
+
+```json
+{
+  "cycles": [
+    {
+      "id": "blind-2027",
+      "venue": { "kind": "preset", "extends": "./blindconf.jsonc" },
+      "opened": "2026-09-09",
+      "phase": "porting"
+    }
+  ]
+}
+```
+
+where the preset's `template` is `\documentclass[conference,compsoc]{IEEEtran}`, and `paper.tex`
+begins with that very line:
+
+```text
+papers/my-paper/paper.tex
+  1:1  error  the \documentclass is already `\documentclass[conference,compsoc]{IEEEtran}`, which
+              BlindConf requires, and the cycle «blind-2027» still declares "phase": "porting" — the
+              port is done: remove the phase, so the venue's format rules judge this paper again
+              cycle/port-done
+```
+
+Passing:
+
+- the same paper with the phase removed (or `"phase": "prepared"`);
+- a cycle still porting whose `paper.tex` still has the old venue's class line — the port is open
+  work, as declared;
+- a preset that names no `template`: there is nothing to compare, as for `tex/template`;
+- a paper whose `paperlint.json` has no `cycles`.
+
+## Options / preset fields
+
+No rule options. It reads the preset's `template`, the field [`tex/template`](../tex/template.md)
+reads, and compares the way that rule does.
+
+## What it does not check
+
+- Whether the port is complete beyond the class line — the page limit, the fonts, the sections. Those
+  are the format rules' to judge, and they judge again the moment the phase is dropped.
+- A cycle declared `porting` whose class line is still the old one: that is the state the phase
+  describes.
+- A `porting` cycle that has already sent a PDF — [`cycle/stage`](stage.md) reports that on the
+  scorecard.
+
+## How to fix
+
+Remove `"phase": "porting"` from the current cycle in `paperlint.json`, then run `paperlint lint`
+and `paperlint build`: the venue's format rules judge the paper again, and what they report is the
+rest of the port.

package/docs/rules/cycle/record.md

@@ -0,0 +1,152 @@
+# cycle/record
+
+**Level:** error · **Reads:** the paper's `paperlint.json` (`cycles`) and the venue presets it names ·
+**Reported on:** the first line of the paper's `PIPELINE-STATUS.md`
+
+## What it catches
+
+The paper's `cycles` — its attempts at venues, one entry each, oldest first — do not describe one
+consistent history:
+
+- **one source's deadlines are out of order.** The order a venue's deadlines keep is
+
+  ```text
+  registration ≤ submission ≤ resubmission < notification < camera-ready
+  ```
+
+  and it is checked within each source: the portal's readings among themselves, the call's among
+  themselves, the overrides among themselves. A call's abstract day after its own paper day is a
+  date typed wrong or an AoE day miscounted.
+
+- **a portal reading the venue's preset already carries.** A cycle whose venue is a preset derives
+  the portal's readings from it (`presets/msr.jsonc` carries MSR's); the same reading typed into the
+  paper is a second copy, and it goes stale the day paperlint updates the preset.
+
+The structural problems of the record — a bare date where an instant is needed, two open cycles, an
+open cycle that is not the last, two cycles with one id, one deadline in two entries, two readings
+from one source, an override without its reason or evidence, an unknown key anywhere in the record
+— are refused earlier, by `paperlint lint` itself before any rule runs, with one line naming the
+file, the place and the problem, and exit code 2, the same way an unknown key in `paperlint.json` is
+refused. This rule judges what the record parses with.
+
+## Why
+
+A deadline is the one fact in the record whose error costs the attempt. The record keeps what each
+page said — the portal, which refuses the upload, and the call, which only announces — with where and
+when it was read, and the **earlier** reading binds unless a human override says otherwise. That the
+portal and the call disagree is expected and is not a finding: at MSR 2027 the portal's "submissions
+must be completed by" falls at 04:00 UTC on the call's _abstract_ day, four days before the call's
+_paper_ day, and keeping both is how the earlier one is seen. What is a finding is a source that
+contradicts itself, and a fact recorded in two places.
+
+## Examples
+
+Failing — `paperlint.json`, the call's own dates inverted:
+
+```json
+{
+  "cycles": [
+    {
+      "id": "conf-2027",
+      "venue": { "kind": "preset", "extends": "paperlint:acm-sigconf" },
+      "opened": "2026-09-09",
+      "deadlines": [
+        {
+          "what": "registration",
+          "observed": [
+            {
+              "at": "2026-10-25 AoE",
+              "source": "call",
+              "url": "https://conf.example/cfp",
+              "read": "2026-09-09"
+            }
+          ]
+        },
+        {
+          "what": "submission",
+          "observed": [
+            {
+              "at": "2026-10-23 AoE",
+              "source": "call",
+              "url": "https://conf.example/cfp",
+              "read": "2026-09-09"
+            }
+          ]
+        }
+      ]
+    }
+  ]
+}
+```
+
+```text
+papers/my-paper/PIPELINE-STATUS.md
+  1:1  error  paperlint.json, cycle «conf-2027»: the call's "registration" (2026-10-26T11:59:59Z) is
+              after its "submission" (2026-10-24T11:59:59Z) — a date typed wrong, or an AoE day
+              miscounted (write a call's date as "YYYY-MM-DD AoE" and paperlint converts it)
+              cycle/record
+```
+
+Failing — a cycle at `paperlint:msr` that copies the portal's reading the preset carries:
+
+```text
+  1:1  error  paperlint.json, cycle «msr-2027»: the "submission" deadline records a portal reading,
+              and the preset paperlint:msr carries the portal's reading of it (2026-10-20T04:00:00Z,
+              read 2026-10-05) — a second copy drifts when paperlint updates the preset: delete this
+              one (the call's reading and an override stay)  cycle/record
+```
+
+Passing:
+
+- the first record with the registration on `2026-10-20 AoE`;
+- a cycle at `paperlint:msr` recording the call's abstract day as `registration` (`2026-10-20 AoE`)
+  and its paper day as `submission` (`2026-10-23 AoE`): the preset's portal `submission`
+  (`2026-10-20T04:00:00Z`) is earlier than both, binds, and is not a contradiction;
+- a cycle with no `deadlines` at all;
+- a paper whose `paperlint.json` has no `cycles` (the flat form, `"extends": "paperlint:…"`).
+
+## Options / preset fields
+
+No rule options. One deadline entry:
+
+```jsonc
+{
+  "what": "submission", // registration (the call's abstract) · submission (the full paper) · resubmission · notification · camera-ready
+  "observed": [
+    {
+      "at": "2026-10-23 AoE", // or an instant with its zone: 2026-10-20T04:00:00Z
+      "source": "call", // or "portal"
+      "url": "https://conf.example/cfp",
+      "read": "2026-10-05", // the day the page was read
+    },
+  ],
+  // optional; wins over every reading
+  "override": {
+    "at": "2026-10-27T04:00:00Z",
+    "reason": "the chairs extended the deadline by mail",
+    "evidence": "mail/extension.eml",
+  },
+}
+```
+
+The preset field is `deadlines` in a venue preset: a list of readings `{ "what", "at", "source",
+"url", "read" }`, `at` in UTC. Shipped presets carry their portal's readings and are refreshed by a
+scheduled workflow in paperlint's repository; your own preset may carry readings the same way.
+
+## What it does not check
+
+- Whether a reading is still what the page says. The rule has no network and no clock; a shipped
+  preset's readings are re-read daily in paperlint's repository and reach you as a patch release.
+  A venue without a preset (`"kind": "named"`) is re-read by you.
+- Whether a deadline has passed, or how long is left.
+- A disagreement between sources — that is what the record holds; the earlier reading binds.
+- Whether an override's evidence exists — that is [`cycle/evidence`](evidence.md).
+- Whether the scorecard's stages name these cycles — that is [`cycle/stage`](stage.md).
+
+## How to fix
+
+Correct the date. For a deadline read from a call for papers, write it as the call prints it,
+`"YYYY-MM-DD AoE"`, and let paperlint do the arithmetic; for one read from a portal, write the instant
+the portal shows with its zone. Delete a portal reading the venue's preset already carries. A
+deadline the chairs moved by mail is an `override`, with the mail saved in the paper folder as its
+`evidence`.

package/docs/rules/cycle/stage.md

@@ -0,0 +1,85 @@
+# cycle/stage
+
+**Level:** error · **Reads:** the paper's `PIPELINE-STATUS.md` frontmatter (`stages`) and its
+`paperlint.json` (`cycles`) · **Reported on:** the frontmatter of `PIPELINE-STATUS.md`
+
+## What it catches
+
+With `cycles` declared in the paper's `paperlint.json`, a `stages` entry in the scorecard's
+frontmatter must say which attempt it belongs to, `cycle: <id>`, and that attempt must exist. Three
+things can be wrong:
+
+- a stage names no `cycle`;
+- a stage names a `cycle` no declared cycle has as its `id`;
+- a `submitted` or `camera-ready` stage belongs to a cycle whose `phase` is still `porting` — a PDF
+  was sent to the venue while the record says the source is still in the previous venue's template.
+
+Without `cycles` (the flat form) the rule is silent and `venue:` stays what it was.
+
+## Why
+
+A paper's attempts are recorded once, in `paperlint.json`; the scorecard's stages are the evidence
+— which PDF, how many bytes, when. A stage that names its cycle says which attempt the evidence
+belongs to, and the venue's name is then spelled once, in the preset, instead of as a free string
+retyped per stage (`REALM @ EMNLP 2026`, `AISec 2026 @ ACM CCS`, …). A stage whose cycle does not
+exist is evidence of nothing.
+
+The `porting` case is the other half of a declaration: `"phase": "porting"` silences the venue's
+format rules because the source is, by declaration, not yet in the venue's template. A PDF sent
+from such a cycle contradicts that declaration — either the port is done and the phase is stale, or
+the stage is wrong.
+
+## Examples
+
+Failing — `paperlint.json` declares one cycle, `open-2027`, and `PIPELINE-STATUS.md` has
+
+```yaml
+---
+stages:
+  - stage: submitted
+    date: 2026-07-22
+    pdf: versions/a.pdf
+    bytes: 1
+---
+```
+
+```text
+papers/my-paper/PIPELINE-STATUS.md
+  1:1  error  the «submitted» stage (2026-07-22) names no `cycle` — this paper declares `cycles` in
+              paperlint.json, so each stage says which attempt it belongs to: `cycle: <id>`, one of
+              open-2027  cycle/stage
+```
+
+Passing:
+
+```yaml
+---
+stages:
+  - stage: submitted
+    cycle: open-2027
+    date: 2026-07-22
+    pdf: versions/a.pdf
+    bytes: 1
+---
+```
+
+Also passing: a paper without `cycles`, whatever its stages say.
+
+## Options / preset fields
+
+No rule options.
+
+## What it does not check
+
+- The PDF, its bytes, or the frozen source — [`paper/stages`](../../rules.md) and `paper/source`
+  do.
+- Whether a stage's `date` falls inside its cycle (after `opened`, before the outcome's `date`).
+- Whether a cycle without any `submitted` stage was in fact submitted; that is the portal's record
+  (`paperlint submission`), not the scorecard's.
+- The rest of the cycle record — [`cycle/record`](record.md) does.
+
+## How to fix
+
+Add `cycle: <id>` to each stage, with the id of the cycle the PDF was sent in. For a stage whose
+cycle is still `porting`: if the port is done, remove `"phase": "porting"` from the cycle
+([`cycle/port-done`](port-done.md) says when); if no PDF was sent, remove the stage.

package/docs/rules/tex/claim-provenance.md

@@ -68,7 +68,6 @@ any other rule.
   3 of the model_) is read as a quantity, and so is the number of a name that opens the sentence
   (_Claude 3 fails …_).
 - Whether a count followed by a plural noun is really the sample (_3 times faster_ passes as one).
-- Markdown papers (`paper.md`): it reads the LaTeX parse tree only.
 
 ## How to fix