feat(ci): govulncheck - add gVisor sandboxing via geomys/sandboxed-step

[kakkoyun]

What does this PR do?

Wraps govulncheck execution steps in geomys/sandboxed-step@v1.2.1 to confine CI execution using gVisor syscall filtering.

This PR will fail CI because geomys/sandboxed-step is not in the DataDog enterprise action allowlist. This PR exists to start the discussion for adding it.

Motivation

Per Filippo Valsorda's article:

"Supply chain attacks have a short half-life! You can further mitigate the risk by using a CI sandboxing mechanism like geomys/sandboxed-step, which uses gVisor to remove the ambient authority that GitHub Actions grants every workflow, including supposedly read-only ones."

govulncheck runs go install and executes module code during analysis — if a dependency contains malicious code, it runs in CI with full ambient authority (network access, token access, filesystem access). The persist-credentials: false change in #4597 removes the checkout-persisted credential, but the runner still has network access and other ambient authority. gVisor sandboxing removes all of this.

To merge this PR: request adding geomys/sandboxed-step to the DataDog enterprise GitHub Actions allowlist.

What geomys/sandboxed-step provides

• Syscall filtering: gVisor intercepts all syscalls, preventing unauthorized file access and process spawning
• Network isolation: prevents exfiltration of secrets, tokens, or source code
• Ambient authority removal: GitHub Actions grants every workflow ambient authority; the sandbox strips it

References

• Filippo Valsorda: Turn Dependabot Off — supply chain attack mitigation section
• geomys/sandboxed-step — gVisor-based CI sandboxing action
• gVisor — application kernel for container isolation
• PR #4595 — parent govulncheck migration PR
• PR #4597 — persist-credentials: false (stacked below this PR)

Reviewer's Checklist

• N/A — CI/workflow-only change, no Go code modified

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 60.58%. Comparing base (31f8814) to head (f5cf6b9).

Additional details and impacted files

see 17 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[datadog-prod-us1-6]

✅ Tests

🎉 All green!

❄️ No new flaky tests detected
🧪 All tests passed

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 60.05% (+3.91%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: aa79d34 | Docs | Datadog PR Page | Was this helpful? React with 👍/👎 or give us feedback!}

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-03-27 11:41:33

Comparing candidate commit aa79d34 in PR branch kakkoyun/govulncheck-sandboxed-step with baseline commit 54c1038 in branch kakkoyun/govulncheck.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 216 metrics, 8 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'