VasanthandClaude Opus 5.5
58baf7f778
On Windows a multi-line argv prompt is cut at its first newline, which also dropped the permission flags. The a-empty eval now passes every check, including the page check on the page the skill built. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Tests
Run everything from the repo root:
uv run --with pytest --with playwright pytest tests
-
test_*.py: assert-based self-checks for template logic (lint of unfilled Python templates, the custom palette, presets in step with the template). -
page_check/: Seam 1. Loads a dashboard page in headless Chromium with no backend, pushes contract envelopes throughwindow.__pyMsg, and asserts on what is visible. Every case runs in light, dark and narrow (800px) mode.- By default it checks the no-AI template, filled with test values. To check a
generated page instead, set
PAGE_CHECK_HTML=path/to/dashboard.html. - The page sees a fake Qt bridge. Every action the page sends is recorded in
window.__callsas[slot, ...args]. - To add a case, write a
test_*.pyfile inpage_check/that uses thepageandpush(event, data)fixtures fromconftest.py.
- By default it checks the no-AI template, filled with test values. To check a
generated page instead, set
-
evals/: Seam 2. Runs the skills on a fixture project withclaude -p, then grades the reply and runs the page check on the page it built:uv run --with pytest --with playwright python tests/evals/run_eval.py a-emptyResults go to
evals/results/<fixture>.json. To add a fixture, add an entry toFIXTURESinrun_eval.py. -
fixtures/<letter>-<name>/: small test projects for the evals.