Release BenchFlow 0.7.7 - #1119
Merged
Merged
Conversation
Publish the `network_mode: denylist` egress policy for Docker and Daytona — a task can keep the internet reachable while the paper it is built from, its mirrors, and its code repository stay off limits — together with the hardening that makes the mode usable across harnesses: the agent's own model gateway stays reachable, every ACP harness gets the same proxy, certificates and UID firewall, codex-acp is configured through native ACP settings, and `cryptography` moves to a core dependency. Set the public package version to 0.7.7.
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
bingran-you
temporarily deployed
to
pypi-internal-preview
September 9, 2026 22:36 — with 
GitHub Actions
Inactive
GitHub Actions
Inactive
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Publish
network_mode: denylistand set the public package version to 0.7.7.What ships
The 0.7.7 line is the URL/host denylist egress policy and the hardening that makes it usable in practice:
network_mode: denyliston Docker and Daytona (feat(sandbox): enforce a URL denylist egress mode on docker and daytona #1113). The task keeps internet access whileblocked_urlsandblocked_hostsbecome unreachable for the agent — the paper a task is built from, its mirrors, and its code repository. Enforced by a root-owned loopback proxy behind the sandbox-user firewall, with hosted search switched off per harness and every refusal logged totrajectory/egress_denylist.jsonl. Other backends refuse the mode at preflight.ProxyAgentignoreNO_PROXY), the policy applies identically to every ACP harness and later sandbox roles,codex-acpis configured through native ACP settings rather than-coverrides, andcryptography>=44moves from thesandbox-agentcoreextra to a core dependency because both backends mint TLS certificates.This PR
Same three-file shape as Release BenchFlow 0.7.6: CHANGELOG heading over the accumulated entries,
pyproject.toml0.7.7.dev0->0.7.7, and the matchinguv.lockline.Verification
uv lock --checkcleanuv run pytest tests/test_release_version.py tests/test_release_workflow_wiring.py tests/test_release_hardening_r5.py— 41 passedTAG_NAME=v0.7.7 tools/release_version.py public-release->version=0.7.7uv build --no-sources+twine check dist/*— both distributions PASSEDAfter merge
Push
v0.7.7to triggerpublic-release.yml(PyPI + GitHub Release), then bumpmainto0.7.8.dev0.