Repository navigation
fix(deps): raise torch-server development floors past known advisories - #111
Merged
Merged
Conversation
…oors The Pantograph guard interop harness lock was stale: Pumas's own dependencies moved after it was generated, so the documented `--offline --locked` run failed before compiling. Re-resolve it minimally and move tokio, zstd, zstd-safe, zstd-sys, h2, crossbeam-epoch and anyhow to the patched versions rust/Cargo.lock already uses. zip 2.4.2 and pyo3 0.23.5 remain because the pinned Pantograph commit requires zip "2" and pyo3 "0.23"; SECURITY.md records why. Scorecard reads each lower bound in torch-server/requirements.txt as the installed version, so torch>=2.1.0 alone reported 22 advisories. Raise the torch, transformers, fastapi and accelerate floors past their known advisories. The file stays a direct-development input. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
|
#110 repins the harness to a Pantograph commit with patched zip and pyo3 and regenerates the same lock, so keep this change to the torch-server development floors. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
[Low impact] The PR appears safe to merge; no actionable regression was found. SummaryUpdates development dependency floors and refreshes the standalone Pantograph test harness lock.
No actionable issues were found. MrScripty explicitly defers the remaining Reviews (1) · Last reviewed commit: "fix(deps): clear scan findings in the in..." · Reviewed by Greptile |
This was referenced Nov 6, 2026
MrScripty
added a commit
that referenced
this pull request
Nov 6, 2026
The audit remediation landed under [Unreleased], but it ships in the untagged 0.8.0. Merge it into the 0.8.0 section, correct the stale dependency line (Vite 8.3.5, zip is 9.4 not 2.6.1, rusqlite 0.41), and note the shared desktop contract package and the test-harness advisory repins. The release notes gain the desktop/runtime hardening, the Rust API removals and Result signatures, the HTTPS-first plaintext sources, and a pointer to the moved-download recovery steps. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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.
Summary
Scorecard's OSV scan reads each
>=bound intorch-server/requirements.txtas the installed version. Sotorch>=2.1.0alone reported 23 advisories in the "Vulnerabilities" alert, and transformers, fastapi and accelerate reported 6 more.This PR raises those floors past the known advisories:
The file is still an unpinned, direct-development input that managed installs never read. A fresh
pip installalready resolves above all of these, so the change only rules out old vulnerable versions.docs/SECURITY.mdnotes why the floors sit where they do.The interop harness lock was in an earlier revision of this PR. I dropped it because #110 regenerates that lock against a patched Pantograph pin.
Verification
I checked every PyPI advisory in the latest Scorecard Vulnerabilities result against its OSV ranges. None affects the new floors: each one is fixed at or below the new floor, or its last affected version is below it.
🤖 Generated with Claude Code