Skip to content

Normalize the trivy-action version comment to the v-prefixed tag - #64

Open
jeremy wants to merge 1 commit into
mainfrom
normalize-trivy-version-comment
Open

Normalize the trivy-action version comment to the v-prefixed tag#64
jeremy wants to merge 1 commit into
mainfrom
normalize-trivy-version-comment

Conversation

@jeremy

@jeremy jeremy commented Aug 20, 2026

Copy link
Copy Markdown
Member

This unblocks the GitHub Actions audit failure on #63.

aquasecurity/trivy-action published both 0.35.0 and v0.35.0 for that release, so our bare # 0.35.0 comment still resolves and main audits clean today. It stopped publishing the bare alias at 0.36.0 — only v0.36.0 exists:

refs/tags/0.35.0
refs/tags/v0.34.0
refs/tags/v0.35.0
refs/tags/v0.36.0

Dependabot carries the existing comment style forward, so on #63 it writes # 0.36.0, and zizmor's ref-version-mismatch — new in 1.29.0, which #63 pulls in by bumping zizmor-action v0.6.0 → v0.6.2 — correctly flags it as "points to unknown ref ... is pointed to by tag v0.36.0". That single medium finding is the only thing failing #63; its other 15 checks pass.

Fixing the style here rather than in the bump is deliberate: dependabot then picks up the v prefix on this and every future trivy bump, instead of the same finding recurring each time. It also avoids pushing to a Dependabot branch — per the note in basecamp/.github's dependabot-sync-actions-comments.yml, any non-Dependabot push to such a PR flips the actor on the retriggered pull_request runs and lifts GitHub's Dependabot sandbox over freshly bumped, unreviewed action pins.

57a97c7e is the same commit under both tags, so the pin is unchanged — comment only. Verified with zizmor --persona=regular (online) 1.29.0 against this tree: no findings.

Once this lands, rebasing #63 should take it green.

trivy-action published both 0.35.0 and v0.35.0, so the bare comment
still resolves. It stopped publishing the bare alias at 0.36.0, where
only v0.36.0 exists -- so dependabot, which carries the existing comment
style forward, writes '# 0.36.0' and zizmor's ref-version-mismatch
(new in 1.29.0) flags it as pointing to an unknown ref. That is what
fails the audit on #63.

Fixing the style here rather than in the bump means dependabot picks up
the v-prefix from now on. 57a97c7e is the same commit under both tags,
so the pin is unchanged.
Copilot AI balanced review requested due to automatic review settings August 20, 2026 09:31
@github-actions

Copy link
Copy Markdown

Sensitive Change Detection (shadow mode)

This PR modifies control-plane files:

  • .github/workflows/security.yml

Shadow mode — this check is informational only. When activated, changes to these paths will require approval from a maintainer.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Normalizes the Trivy action’s version comment to match its v-prefixed tag without changing the pinned commit.

Changes:

  • Updates 0.35.0 to v0.35.0, preventing future audit mismatches.

Tip

If you aren't ready for review, convert to a draft PR.
Click "Convert to draft" or run gh pr ready --undo.
Click "Ready for review" or run gh pr ready to reengage.


💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

@jeremy

jeremy commented Aug 20, 2026

Copy link
Copy Markdown
Member Author

Heads up on the red Govulncheck here — it is not from this change, which is one character inside a YAML comment.

It fails on Go standard library advisories (Fixed in: encoding/asn1@go1.26.6, 3 stdlib vulns reachable, e.g. editor/editor.go:39:21exec.Commandasn1.Unmarshal). Nothing in a workflow comment can reach Go code analysis.

Confirmed it's time-based rather than tree-based: #63's Govulncheck passed at ~09:24 today, and re-running that same job on the same commit an hour later now fails identically. So a new stdlib advisory landed in between and every open cli PR will show it until the Go toolchain is bumped. Worth its own issue — it's a real finding, just not one this PR introduced or should carry.

GitHub Actions audit — the check this PR actually targets — passes here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants