Experience-Driven Black-Box Cost Control for Coding Agents
Python agentic-aiagentic-codingai-codingai-coding-assistantanthropicartificial-intelligenceclaude-codecodexcoding-agentcoding-agentscost-optimizationcost-reductiondeveloper-toolsgithub-copilotllmopen-sourceopenaiproductivitytoken-optimization
CI: test on Ubuntu; report failing tests as an annotation
Aarond0675a338594
diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.ymlindex d69d03b..8e84c7c 100644--- a/.github/workflows/tests.yml+++ b/.github/workflows/tests.yml@@ -13,17 +13,24 @@ jobs:strategy:fail-fast: falsematrix:- os: [ubuntu-latest, windows-latest]python: ["3.10", "3.13"]- runs-on: ${{ matrix.os }}- defaults:- run:- # Git Bash on Windows: the launchers are `sh` scripts, run there as Claude Code runs them.- shell: bash+ runs-on: ubuntu-lateststeps:- uses: actions/checkout@v5- uses: actions/setup-python@v6with:python-version: ${{ matrix.python }}- run: python -m pip install pytest- - run: python -m pytest tests -rs+ # On failure, the summary lines also go out as an annotation, which the check run shows+ # without signing in.+ - run: |+ set -o pipefail+ python -m pytest tests -rfEs --tb=line 2>&1 | tee pytest.log || {+ status=$?+ msg=$(grep -E '^(FAILED|ERROR) |\.py:[0-9]+: ' pytest.log | head -60 \+ | sed 's/%/%25/g' | awk '{printf "%s%%0A", $0}')+ echo "::error title=pytest::${msg}"+ exit $status+ }+ env:+ COLUMNS: "200"