epsiloneridani-progress 0.1.0__tar.gz

epsiloneridani_progress-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "{}"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright {yyyy} {name of copyright owner}
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

epsiloneridani_progress-0.1.0/PKG-INFO

@@ -0,0 +1,174 @@
+Metadata-Version: 2.4
+Name: epsiloneridani-progress
+Version: 0.1.0
+Summary: Decide, generate, and publish per-roadmap progress reports for EpsilonEridani.
+License: Apache-2.0
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Dynamic: license-file
+
+# EpsilonEridaniProgress
+
+Progress reporting for [EpsilonEridani](https://github.com/eic/EpsilonEridani): what has actually
+been achieved on each roadmap, written for a human to read in a minute.
+
+Each roadmap directory in
+[EpsilonEridaniRoadmaps](https://github.com/eic/EpsilonEridaniRoadmaps) carries two generated files:
+
+- **`STATUS.md`** — a snapshot, rewritten whole on each update. It says which parts of the roadmap
+  are done and sketches the frontier, headed by the commit it describes.
+- **`PROGRESS.md`** — an append-only log. Each section covers one window of merged PRs as a few
+  holistic paragraphs, emphasising named theorems rather than listing every PR.
+
+New `PROGRESS.md` sections are announced in the **EpsilonEridani > Progress logs** Zulip topic.
+
+## Why this repo exists
+
+The rubrics-and-machinery split of
+[EpsilonEridaniReview](https://github.com/eic/EpsilonEridaniReview), applied to reporting: the
+prompts and the tooling live here, the output lands in EpsilonEridaniRoadmaps, and
+[EpsilonEridaniWorker](https://github.com/eic/EpsilonEridaniWorker) drives it.
+
+The design rule is that **a model only ever writes prose**. Every decision — whether an update is
+due, which roadmap it covers, which PRs are in the window, and what mathematics actually landed —
+is made by tested Python before any model starts, and the git and pull-request work afterwards is
+done by tested Python too.
+
+## The commands
+
+```
+epsiloneridani-progress due                      is an update due? (one API call, no clone)
+epsiloneridani-progress plan   --roadmap-dir DIR pick the roadmap and the PR window
+epsiloneridani-progress facts  --plan FILE       what declarations actually landed in the window
+epsiloneridani-progress apply  --plan FILE ...   write the files, open the PR (resumable)
+epsiloneridani-progress announce --section FILE  post a new section to Zulip (idempotent)
+```
+
+`due` is the only one that runs often; it exits 75 ("no progress") when nothing is due, matching
+the worker's convention. `plan` runs at most once a day.
+
+## The window cursor is a SHA, on the docs-tracking branch
+
+A window is the half-open commit range `(from_sha, to_sha]` on EpsilonEridani's **`docgen`** branch, where
+`from_sha` is the `to_sha` of the previous `PROGRESS.md` section.
+
+`docgen` nominates the most recent commit on `main` whose API documentation has been published, and
+the window ends at **the commit the published documentation actually reports** — read from the site
+itself, since the deploy is independent and the branch can sit ahead of it. Ending at the branch tip
+instead would record a cursor covering work the report never described, and because the next window
+starts after that cursor, the work in between would never be reported at all.
+
+The cost is latency: a report describes the project as of the last published docs build rather than
+the tip. That is the right trade for a document whose whole purpose is to be read, and the header
+records the exact commit, so nothing is misdated.
+
+Links are not computed by this project at all, and neither are declaration names. Both are read from
+doc-gen4's own published output — the declaration index and each module page, which carry the exact
+name, kind, source file and line range — so a link resolves because it was read from the page it
+points at. `git blame` over those line ranges is what decides whether a declaration belongs to the
+window.
+
+There is deliberately no Lean parser here. Qualifying a name correctly means resolving `namespace`
+against `section`, `end`, `open ... in` and `_root_`, and many real declarations (projections,
+constructors, `deriving` output) are never written in the source at all. An approximation gets most
+names right, which is the worst outcome available: the wrong ones are indistinguishable from the
+right ones, and a link built from a wrong name is a plausible dead link. PR numbers come from the squash-merge commit
+subjects in that range and are attributed by their `roadmap/<Area>` label.
+
+Wall-clock time is used only for display. A cursor made of timestamps would be wrong: a worker
+clock running fast advances it past PRs whose merge times then fall *before* the stored cursor, and
+those PRs are never reported at all.
+
+## The coverage header
+
+A status snapshot may carry a second machine header beside `epsiloneridani-status:v1`:
+
+```text
+<!--epsiloneridani-status:v1 {"roadmap":"EllipticCurves","to_sha":"…","ts":"…"}-->
+<!--epsiloneridani-coverage:v1 {"layers":[{"id":"Layer 0","remaining":"Weil reciprocity","state":"partial"},{"id":"Layer 1","state":"done"},…],"readme_sha":"…","roadmap":"EllipticCurves","to_sha":"…"}-->
+# Status: EllipticCurves
+```
+
+It is the report's verdict on each layer of the roadmap, in a form a script can read; the prose
+says the same things, but nothing can aggregate prose across forty roadmaps. The consumer is the
+EpsilonEridani site's Progress page (`scripts/roadmap_progress.py` in the EpsilonEridani repository).
+
+**Wire schema.** `roadmap` is the status header's area, or `Area/Child` for a sub-roadmap (below);
+`to_sha` equals the status header's; `readme_sha` is the SHA-256 of the README the layers were read
+from; `layers` is a non-empty list (at most 64) of `{"id", "state", "remaining"?}` with `id` a short
+label (`Layer 3`, `Lane G`, `L0A`), `state` one of `done`, `partial`, `untouched`, `unassessed` (the
+material said nothing), and `remaining` an optional one-line note of at most 200 characters with no
+angle brackets. No other keys anywhere. Keys are sorted and the JSON is compact;
+`files.require_coverage` is the one definition, and both the worker and the gate run it.
+
+**Trust boundary.** The model supplies only the assessments, as JSON inside a fenced
+```` ```coverage ```` block at the end of its status prose. Code supplies everything else: `plan`
+records the README's layer headings (`Layer` / `Lane` / `Part` / `Stage`, or `L0A`-style labels)
+and `readme_sha` in the plan; `apply` removes the block, validates the payload against the schema,
+checks it names every listed layer exactly once and nothing else, and writes the header; the gate
+treats the headers as part of the canonical prefix (the lines after the status header, nowhere else)
+and re-validates them. The gate proves the header's shape, never its truth, and cannot check
+`readme_sha` (it never checks out the roadmap repository): the header is a claim, not a certificate.
+
+**Missing block.** A README with no layer headings gives a plan with no layers and a report with no
+header. When the plan lists layers, the worker refuses a body with no block, just as it refuses a
+block that is malformed, not JSON, or does not fit the plan: a new report changes the report hash,
+which retires the site's hand transcription of the old report, so publishing it headerless would
+turn assessed layers into unassessed ones. The model always has `unassessed` for a layer it cannot
+judge. The gate is deliberately more lenient and still accepts a status file without the header,
+which is what every report written before the header existed looks like.
+
+**README hash.** `readme_sha` binds an assessment to the specification it was made against: layer
+ids alone do not, since a layer's requirements can change under an unchanged heading. The consumer
+refuses a header whose hash does not match the README it reads and shows those layers as
+unassessed with that reason.
+
+**Umbrella areas.** An area whose directories include sub-roadmaps (RepresentationTheory's twelve)
+is one labelled area with one report, so that report is the only account of its children. A
+sub-roadmap is a directory directly below the area with a `README.md` and a `Suggested.lean`, the
+consumer's rule. `plan` lists each one with layer headings under `sub_roadmaps`, with its
+`Area/Child` id, its README path and its own `readme_sha`; the model reads those READMEs and answers
+with a JSON object keyed by roadmap id instead of an array; and `apply` writes one coverage header
+per sub-roadmap, each bound to its own README, after the area's own header (if its README has
+layers) and in ascending order of name. The gate holds each to the same schema, requires
+`Area/Child` with an alphanumeric child directly below the status file's own area, caps them at 32,
+and treats all of them as part of the canonical prefix. A consumer that does not know sub-roadmap
+headers ignores them: their `roadmap` never equals a row's own name.
+
+**Rollout.** Merge here, then bump the two pins in EpsilonEridaniRoadmaps's `progress-*.yml` workflows and
+the worker's `PROGRESS_REF` to the same SHA, together: the generator and the gate must run one
+version. Reports written before the bump simply have no header.
+
+**Leaving the hand transcriptions.** Today the site reads layer states from a hand-transcribed
+file in the EpsilonEridani repository (`scripts/roadmap_coverage.json`), each entry bound to the exact
+report it was read from. Nothing is backfilled: an area's transcription stays valid until its next
+report, which after the bump carries the headers (the worker refuses one without them when there
+are layers), and the site then reads those instead and the transcription is retired. That holds
+for the RepresentationTheory children too, all at once, at the umbrella's next report, provided
+the site reads sub-roadmap headers by then. Each entry can be deleted from the transcription file
+once its roadmap's report carries a header.
+
+## Trust boundary
+
+`STATUS.md` and `PROGRESS.md` are **machine-owned, and their prose is not security-validated**.
+
+The merge gate proves the *shape* of a generated update — its paths, its cursor, that it is a
+byte-exact append, and that the head being merged is the head that was validated. It cannot prove
+that the prose is true. Anyone may open a EpsilonEridani PR whose description contains prompt-injection
+text; once that PR merges legitimately, its description reaches the writing model.
+
+The mitigations reduce the risk and are not claimed to remove it: the model is grounded in
+mechanically-extracted declaration names rather than author prose, PR bodies are delimited and
+size-capped, reserved `epsiloneridani-*:v1` markers are rejected in model output, and Zulip mentions are
+defused. Read these two files as a machine's summary, not as reviewed roadmap content.
+
+**The blast radius is two markdown files AND a Zulip message.** Every merged section is posted to
+**EpsilonEridani > Progress logs** automatically, so accepted prose reaches an audience outside the
+repository. The post is treated as data -- mentions and bare `#123` linkifiers are defused, the
+message is size-capped, it links both the appended log and the current roadmap status, and it is
+idempotent on a stable per-window id -- but it is a second sink and the threat model has to say so.
+
+## Licence
+
+Apache-2.0.

epsiloneridani_progress-0.1.0/README.md

@@ -0,0 +1,164 @@
+# EpsilonEridaniProgress
+
+Progress reporting for [EpsilonEridani](https://github.com/eic/EpsilonEridani): what has actually
+been achieved on each roadmap, written for a human to read in a minute.
+
+Each roadmap directory in
+[EpsilonEridaniRoadmaps](https://github.com/eic/EpsilonEridaniRoadmaps) carries two generated files:
+
+- **`STATUS.md`** — a snapshot, rewritten whole on each update. It says which parts of the roadmap
+  are done and sketches the frontier, headed by the commit it describes.
+- **`PROGRESS.md`** — an append-only log. Each section covers one window of merged PRs as a few
+  holistic paragraphs, emphasising named theorems rather than listing every PR.
+
+New `PROGRESS.md` sections are announced in the **EpsilonEridani > Progress logs** Zulip topic.
+
+## Why this repo exists
+
+The rubrics-and-machinery split of
+[EpsilonEridaniReview](https://github.com/eic/EpsilonEridaniReview), applied to reporting: the
+prompts and the tooling live here, the output lands in EpsilonEridaniRoadmaps, and
+[EpsilonEridaniWorker](https://github.com/eic/EpsilonEridaniWorker) drives it.
+
+The design rule is that **a model only ever writes prose**. Every decision — whether an update is
+due, which roadmap it covers, which PRs are in the window, and what mathematics actually landed —
+is made by tested Python before any model starts, and the git and pull-request work afterwards is
+done by tested Python too.
+
+## The commands
+
+```
+epsiloneridani-progress due                      is an update due? (one API call, no clone)
+epsiloneridani-progress plan   --roadmap-dir DIR pick the roadmap and the PR window
+epsiloneridani-progress facts  --plan FILE       what declarations actually landed in the window
+epsiloneridani-progress apply  --plan FILE ...   write the files, open the PR (resumable)
+epsiloneridani-progress announce --section FILE  post a new section to Zulip (idempotent)
+```
+
+`due` is the only one that runs often; it exits 75 ("no progress") when nothing is due, matching
+the worker's convention. `plan` runs at most once a day.
+
+## The window cursor is a SHA, on the docs-tracking branch
+
+A window is the half-open commit range `(from_sha, to_sha]` on EpsilonEridani's **`docgen`** branch, where
+`from_sha` is the `to_sha` of the previous `PROGRESS.md` section.
+
+`docgen` nominates the most recent commit on `main` whose API documentation has been published, and
+the window ends at **the commit the published documentation actually reports** — read from the site
+itself, since the deploy is independent and the branch can sit ahead of it. Ending at the branch tip
+instead would record a cursor covering work the report never described, and because the next window
+starts after that cursor, the work in between would never be reported at all.
+
+The cost is latency: a report describes the project as of the last published docs build rather than
+the tip. That is the right trade for a document whose whole purpose is to be read, and the header
+records the exact commit, so nothing is misdated.
+
+Links are not computed by this project at all, and neither are declaration names. Both are read from
+doc-gen4's own published output — the declaration index and each module page, which carry the exact
+name, kind, source file and line range — so a link resolves because it was read from the page it
+points at. `git blame` over those line ranges is what decides whether a declaration belongs to the
+window.
+
+There is deliberately no Lean parser here. Qualifying a name correctly means resolving `namespace`
+against `section`, `end`, `open ... in` and `_root_`, and many real declarations (projections,
+constructors, `deriving` output) are never written in the source at all. An approximation gets most
+names right, which is the worst outcome available: the wrong ones are indistinguishable from the
+right ones, and a link built from a wrong name is a plausible dead link. PR numbers come from the squash-merge commit
+subjects in that range and are attributed by their `roadmap/<Area>` label.
+
+Wall-clock time is used only for display. A cursor made of timestamps would be wrong: a worker
+clock running fast advances it past PRs whose merge times then fall *before* the stored cursor, and
+those PRs are never reported at all.
+
+## The coverage header
+
+A status snapshot may carry a second machine header beside `epsiloneridani-status:v1`:
+
+```text
+<!--epsiloneridani-status:v1 {"roadmap":"EllipticCurves","to_sha":"…","ts":"…"}-->
+<!--epsiloneridani-coverage:v1 {"layers":[{"id":"Layer 0","remaining":"Weil reciprocity","state":"partial"},{"id":"Layer 1","state":"done"},…],"readme_sha":"…","roadmap":"EllipticCurves","to_sha":"…"}-->
+# Status: EllipticCurves
+```
+
+It is the report's verdict on each layer of the roadmap, in a form a script can read; the prose
+says the same things, but nothing can aggregate prose across forty roadmaps. The consumer is the
+EpsilonEridani site's Progress page (`scripts/roadmap_progress.py` in the EpsilonEridani repository).
+
+**Wire schema.** `roadmap` is the status header's area, or `Area/Child` for a sub-roadmap (below);
+`to_sha` equals the status header's; `readme_sha` is the SHA-256 of the README the layers were read
+from; `layers` is a non-empty list (at most 64) of `{"id", "state", "remaining"?}` with `id` a short
+label (`Layer 3`, `Lane G`, `L0A`), `state` one of `done`, `partial`, `untouched`, `unassessed` (the
+material said nothing), and `remaining` an optional one-line note of at most 200 characters with no
+angle brackets. No other keys anywhere. Keys are sorted and the JSON is compact;
+`files.require_coverage` is the one definition, and both the worker and the gate run it.
+
+**Trust boundary.** The model supplies only the assessments, as JSON inside a fenced
+```` ```coverage ```` block at the end of its status prose. Code supplies everything else: `plan`
+records the README's layer headings (`Layer` / `Lane` / `Part` / `Stage`, or `L0A`-style labels)
+and `readme_sha` in the plan; `apply` removes the block, validates the payload against the schema,
+checks it names every listed layer exactly once and nothing else, and writes the header; the gate
+treats the headers as part of the canonical prefix (the lines after the status header, nowhere else)
+and re-validates them. The gate proves the header's shape, never its truth, and cannot check
+`readme_sha` (it never checks out the roadmap repository): the header is a claim, not a certificate.
+
+**Missing block.** A README with no layer headings gives a plan with no layers and a report with no
+header. When the plan lists layers, the worker refuses a body with no block, just as it refuses a
+block that is malformed, not JSON, or does not fit the plan: a new report changes the report hash,
+which retires the site's hand transcription of the old report, so publishing it headerless would
+turn assessed layers into unassessed ones. The model always has `unassessed` for a layer it cannot
+judge. The gate is deliberately more lenient and still accepts a status file without the header,
+which is what every report written before the header existed looks like.
+
+**README hash.** `readme_sha` binds an assessment to the specification it was made against: layer
+ids alone do not, since a layer's requirements can change under an unchanged heading. The consumer
+refuses a header whose hash does not match the README it reads and shows those layers as
+unassessed with that reason.
+
+**Umbrella areas.** An area whose directories include sub-roadmaps (RepresentationTheory's twelve)
+is one labelled area with one report, so that report is the only account of its children. A
+sub-roadmap is a directory directly below the area with a `README.md` and a `Suggested.lean`, the
+consumer's rule. `plan` lists each one with layer headings under `sub_roadmaps`, with its
+`Area/Child` id, its README path and its own `readme_sha`; the model reads those READMEs and answers
+with a JSON object keyed by roadmap id instead of an array; and `apply` writes one coverage header
+per sub-roadmap, each bound to its own README, after the area's own header (if its README has
+layers) and in ascending order of name. The gate holds each to the same schema, requires
+`Area/Child` with an alphanumeric child directly below the status file's own area, caps them at 32,
+and treats all of them as part of the canonical prefix. A consumer that does not know sub-roadmap
+headers ignores them: their `roadmap` never equals a row's own name.
+
+**Rollout.** Merge here, then bump the two pins in EpsilonEridaniRoadmaps's `progress-*.yml` workflows and
+the worker's `PROGRESS_REF` to the same SHA, together: the generator and the gate must run one
+version. Reports written before the bump simply have no header.
+
+**Leaving the hand transcriptions.** Today the site reads layer states from a hand-transcribed
+file in the EpsilonEridani repository (`scripts/roadmap_coverage.json`), each entry bound to the exact
+report it was read from. Nothing is backfilled: an area's transcription stays valid until its next
+report, which after the bump carries the headers (the worker refuses one without them when there
+are layers), and the site then reads those instead and the transcription is retired. That holds
+for the RepresentationTheory children too, all at once, at the umbrella's next report, provided
+the site reads sub-roadmap headers by then. Each entry can be deleted from the transcription file
+once its roadmap's report carries a header.
+
+## Trust boundary
+
+`STATUS.md` and `PROGRESS.md` are **machine-owned, and their prose is not security-validated**.
+
+The merge gate proves the *shape* of a generated update — its paths, its cursor, that it is a
+byte-exact append, and that the head being merged is the head that was validated. It cannot prove
+that the prose is true. Anyone may open a EpsilonEridani PR whose description contains prompt-injection
+text; once that PR merges legitimately, its description reaches the writing model.
+
+The mitigations reduce the risk and are not claimed to remove it: the model is grounded in
+mechanically-extracted declaration names rather than author prose, PR bodies are delimited and
+size-capped, reserved `epsiloneridani-*:v1` markers are rejected in model output, and Zulip mentions are
+defused. Read these two files as a machine's summary, not as reviewed roadmap content.
+
+**The blast radius is two markdown files AND a Zulip message.** Every merged section is posted to
+**EpsilonEridani > Progress logs** automatically, so accepted prose reaches an audience outside the
+repository. The post is treated as data -- mentions and bare `#123` linkifiers are defused, the
+message is size-capped, it links both the appended log and the current roadmap status, and it is
+idempotent on a stable per-window id -- but it is a second sink and the threat model has to say so.
+
+## Licence
+
+Apache-2.0.