Serve /prototype/lane2 (#659) #674

Merged
claude-bot merged 1 commit from fix/659-lane2-route into main 2026-09-16 18:03:12 +00:00
Collaborator

#673 added static/prototype/lane2.html and a link to it from the prototype index, but not the route that serves it. /prototype/lane2 was a 404 and the new link went nowhere.

Each lane page gets an explicit route in app.py — there is no directory listing for static/prototype/ — so adding the file is only half of adding a page. Added, and added to the parametrised route test beside lane1.

On the testing

CI does not run tools/feedback-site/tests at all. .forgejo/workflows/ci.yml has jobs for the backend, the bot, the frontend, the evals and the Docker build; the feedback site is not among them, and its README documents a manual docker run instead. So nothing in the pipeline would have caught this, and nothing will catch the next one.

Docker Desktop is down on this machine, so I ran them in a scratch venv instead: 80 passed, including the two new parametrisations (GET and HEAD on /prototype/lane2). Caveat worth stating: that venv is Python 3.14 with unpinned deps, not the documented python:3.12-slim. For a route-resolution test that difference does not matter, but it is not the documented command.

Filed separately: the CI gap itself.

🤖 Generated with Claude Code

#673 added `static/prototype/lane2.html` and a link to it from the prototype index, but not the route that serves it. `/prototype/lane2` was a 404 and the new link went nowhere. Each lane page gets an explicit route in `app.py` — there is no directory listing for `static/prototype/` — so adding the file is only half of adding a page. Added, and added to the parametrised route test beside `lane1`. ## On the testing **CI does not run `tools/feedback-site/tests` at all.** `.forgejo/workflows/ci.yml` has jobs for the backend, the bot, the frontend, the evals and the Docker build; the feedback site is not among them, and its README documents a manual `docker run` instead. So nothing in the pipeline would have caught this, and nothing will catch the next one. Docker Desktop is down on this machine, so I ran them in a scratch venv instead: **80 passed**, including the two new parametrisations (`GET` and `HEAD` on `/prototype/lane2`). Caveat worth stating: that venv is Python 3.14 with unpinned deps, not the documented `python:3.12-slim`. For a route-resolution test that difference does not matter, but it is not the documented command. Filed separately: the CI gap itself. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
fix(feedback-site): serve /prototype/lane2 (#659)
All checks were successful
CI / Backend lint (ruff) (pull_request) Successful in 55s
CI / Docker image build (pull_request) Successful in 16s
CI / Bot/backend version sync (pull_request) Successful in 30s
CI / Summarisation accuracy eval harness (stub provider) (pull_request) Successful in 1m43s
CI / Bot tests and audit (pull_request) Successful in 2m39s
CI / Frontend tests, audit, and build (pull_request) Successful in 3m6s
CI / Backend migration, tests, and audit (pull_request) Successful in 10m4s
CI / Synthetic session harness (no GPU, no LLM) (pull_request) Successful in 12m50s
7b236aa20e
Lane 2's page was added without its route, so /prototype/lane2 was a 404
and the link the prototype index had just gained went nowhere. Each lane
page gets an explicit route here rather than a directory listing, so a
new page is not reachable until one is added.

Added to the route test alongside lane1.

Worth noting: CI does not run tools/feedback-site/tests at all, so
nothing would have caught this. Ran them locally instead — 80 passed.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
claude-bot deleted branch fix/659-lane2-route 2026-09-16 18:03:13 +00:00
Sign in to join this conversation.
No description provided.