Handle single-file env import

[willccbb]

Description

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Test improvement

Testing

• All existing tests pass when running uv run pytest locally.
• New tests have been added to cover the changes

Checklist

• My code follows the style guidelines of this project as outlined in AGENTS.md
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Additional Notes

---

Note

Medium Risk
Touches eval CLI configuration resolution; incorrect module path detection could silently skip defaults or pick the wrong pyproject.toml, changing evaluation run parameters.

Overview
prime eval run now loads [tool.verifiers.eval] defaults (num_examples, rollouts_per_example) from an environment’s pyproject.toml for both package and single-file environment modules by switching get_env_eval_defaults to resolve module paths via importlib.util.find_spec.

Adds pytest coverage for both module shapes, and updates the environments docs to include vf.TunnelError under vf.InfraError (plus a small import cleanup in HybridMathRubric).

^{Written by Cursor Bugbot for commit c59f487. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

[cursor]

Skill docs missing for workflow change

Low Severity

This PR changes evaluation workflow behavior in verifiers/scripts/eval.py by altering how get_env_eval_defaults resolves pyproject.toml for environments, but no corresponding skills/*/SKILL.md updates are included. That violates the Skills Updates rule for user-facing workflow changes in verifiers/scripts/*.py.

 

^{Triggered by project rule: BugBot Instructions}

[willccbb]

@cursoragent fix merge conflicts vs main

[cursor]

• Merged origin/main into codex/treat-singlefile-envs-as-packages and resolved the conflict in verifiers/envs/experimental/cli_agent_env.py by standardizing on the safer _require_interception_server() access pattern.
• Ran linting via uvx (ruff) with no outstanding issues.
• Attempted to run tests; several integration tests require external credentials (e.g., PRIME_API_KEY) and failed locally. Since your request was to fix merge conflicts, I did not alter tests or add secrets.

All changes have been pushed to the branch.

• Key edit: ensured tunnel and interception access consistently use _require_interception_server() in CliAgentEnv to avoid None dereferences.