chore(deps): unblock CI — bundle security bumps, docker/action updates, fix ruff format - #226
Merged
Conversation
…s, fix ruff format Consolidates 10 stalled Renovate PRs (#215–#225) that were each individually BLOCKED by three systematic problems on main. No single PR could go green because the Security (pip-audit) job requires all vulnerable deps fixed together, and the Lint job failed for every PR. Root causes fixed: - Lint & Format: ruff 0.16.3 reformats Python code blocks inside markdown; 3 docs (Development.md, Niles-Core-Spec.md, Quality-Assessment.md) were unformatted. - Security / Trivy: mcp 1.27.0, cryptography 49.0.0, json-repair 0.59.5 had HIGH advisories (PYSEC-2026-3481/82/83, CVE-2026-69247, GHSA-xf7x-x43h-rpqh). Bumped to the security-fix versions (mcp 1.28.1, cryptography 50.0.0, json-repair 0.60.1). Verified: fresh image Trivy scan is now clean (0 CRITICAL/HIGH); OS-package HIGHs are already handled by the existing apt-get upgrade in the runtime stage. Bundled version updates (supersedes the individual Renovate PRs): - langfuse docker tag v3 -> v4 (#224) - prom/prometheus v3.13.0 -> v3.13.2 (#216) - ghcr.io/astral-sh/uv 0.11.28 -> 0.12.5 (#218) - actions/checkout v7.0.0 -> v7.0.1 across all workflows (#221, #222) - actions/setup-python v6 -> v7 (#220) - anthropics/claude-code-action digest bump (#215) - mcp/cryptography/json-repair security bumps (#217, #219, #225) Also fixes a marker bug in scripts/test.sh: it excluded llm_judge but not llm_eval, so live-Ollama eval tests ran (and hung) on dev machines. Now aligned with CI. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This was referenced Aug 18, 2026
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
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.
Warum
Alle 10 offenen Renovate-PRs (#215–#225) waren einzeln
BLOCKED. Ursache waren drei systematische Probleme aufmain, nicht die einzelnen Updates — kein einzelner PR konnte grün werden, weil der Security-Job (pip-audit) alle Vuln-Deps gemeinsam gefixt braucht und der Lint-Job für jeden PR fehlschlug.Dieser PR bündelt alles in einen grünen Change.
Root Causes gefixt
ruff 0.16.3reformatiert Python-Codeblöcke in Markdown → 3 Docs (Development.md,Niles-Core-Spec.md,Quality-Assessment.md) neu formatiertmcp 1.27.0→1.28.1,cryptography 49.0.0→50.0.0,json-repair 0.59.5→0.60.1(PYSEC-2026-3481/82/83, CVE-2026-69247, GHSA-xf7x-x43h-rpqh)apt-get upgradeim Runtime-Stage abgedeckt — lokal mit frischem--no-cacheBuild verifiziertLokale Verifikation: frischer Image-Build +
trivy image --severity CRITICAL,HIGH --ignore-unfixed→ 0 Findings.ruff check/formatgrün. Volle Unit-Suite: 1129 passed, 129 deselected.Gebündelte Version-Updates (supersedes)
v3 → v4— chore(deps): update langfuse/langfuse docker tag to v4 - autoclosed #224prom/prometheus v3.13.0 → v3.13.2— chore(deps): update prom/prometheus docker tag to v3.14.0 #216ghcr.io/astral-sh/uv 0.11.28 → 0.12.5— chore(deps): update ghcr.io/astral-sh/uv docker tag to v0.12.5 - autoclosed #218actions/checkout v7.0.0 → v7.0.1(alle Workflows) — chore(deps): update actions/checkout digest to 3d3c42e - autoclosed #221, chore(deps): update actions/checkout action to v7.0.1 - autoclosed #222actions/setup-python v6 → v7— chore(deps): update actions/setup-python action to v7 - autoclosed #220anthropics/claude-code-actiondigest bump — chore(deps): update anthropics/claude-code-action digest to 459ad35 #215Zusätzlich
Fix eines Marker-Bugs in
scripts/test.sh: schlossllm_judgeaus, aber nichtllm_eval→ Live-Ollama-Evals liefen (und hingen) auf Dev-Maschinen. Jetzt konsistent mit CI.🤖 Generated with Claude Code