Build and run agents you can see, understand and trust.
Python agentchatbotlarge-language-modelsllmllm-agentmcpmulti-agentmulti-modalreact-agentrealtime-agentvoice-agentvoice-assistant
fix(ci): let the triage workflows create their environment (#2680)
DavdGao5ff52f877de1
diff --git a/.github/workflows/issue-claim.yml b/.github/workflows/issue-claim.ymlindex c98d198f1..37cbbefe0 100644--- a/.github/workflows/issue-claim.yml+++ b/.github/workflows/issue-claim.yml@@ -71,8 +71,8 @@ jobs:env:DASHSCOPE_API_KEY: ${{ secrets.DASHSCOPE_API_KEY }}run: |- uv venv --python 3.11- uv pip install -e ".[dev]"+ uv venv --clear --python 3.11+ uv pip install --python .venv/bin/python -e ".[dev]"result=$(.venv/bin/python scripts/claim_intent.py \--comment-file /tmp/claim/comment.md \--model "$TRIAGE_MODEL")diff --git a/.github/workflows/issue-triage.yml b/.github/workflows/issue-triage.ymlindex 794b3f92a..a1a7e8531 100644--- a/.github/workflows/issue-triage.yml+++ b/.github/workflows/issue-triage.yml@@ -53,13 +53,16 @@ jobs:- uses: astral-sh/setup-uv@v5with:python-version: '3.11'+ # There is no uv.lock, so without this the cache key never+ # changes and the cache is never invalidated.+ cache-dependency-glob: pyproject.toml# A separate step so a dependency failure is obvious in the log and# costs no model tokens. The agent never installs anything itself.- name: Prepare the environment the agent verifies againstrun: |- uv venv --python 3.11- uv pip install -e ".[dev]"+ uv venv --clear --python 3.11+ uv pip install --python .venv/bin/python -e ".[dev]"# Written as files, never spliced into a shell command: an issue title is# attacker-controlled and `${{ }}` would interpolate it before bash runs.