What does this skill do, and when should you use it?
This skill is dedicated to updating Codex’s pinned V8 and rusty_v8 versions. It follows the repository’s V8 release procedure and edits the relevant Cargo, Bazel, documentation, and checksum surfaces. It prefers checking the hosted v8-canary result, while supporting a clearly identified local substitute when CI is unavailable or local validation is requested. When validation fails, it traces build-relevant upstream changes into Codex’s build graph.
Reads third_party/v8/README.md; inspects and updates codex-rs/Cargo.toml, codex-rs/Cargo.lock, MODULE.bazel, third_party/v8/BUILD.bazel, third_party/v8/README.md, and the matching rusty_v8 SHA-256 manifest; runs the supplied Python checksum commands and unittest suite; checks v8-canary through GitHub check tooling or gh; and, on failure, compares denoland/rusty_v8 with upstream V8 at the target version, examining generated bindings, archive naming, GN/Bazel targets, toolchain inputs, patches, sandboxing, and pointer-compression relationships.
- A Codex maintainer needs to bump the repository’s pinned V8 version.
- A Rust or Bazel engineer needs to refresh rusty_v8 artifacts and their checksum manifest.
- A release owner needs to validate a V8 release candidate on a branch or pull request.
- An engineer needs to diagnose why a V8 version update broke a canary or local build.
How do you install this skill?
- No commands were executed; the scripts, tests, and canary flow were not confirmed to run at this revision.
- Before editing pins, confirm user authorization, preserve the current state, and define recovery steps for failed changes.
- Core validation depends on GitHub/gh and upstream repository access; if unreachable from mainland China, use a local substitute and state its coverage clearly.
- The skill does not explain how to handle credentials, environment variables, or sensitive data in CI logs.
- Shell / CLI
- Network access
- Local filesystem
Python 3GitHub CLI (gh)
The repository places this skill at .codex/skills/update-v8-version/SKILL.md and does not document separate installation for the individual skill. The README documents installation of the Codex CLI collection: on macOS/Linux run curl -fsSL https://chatgpt.com/codex/install.sh | sh; on Windows run powershell -ExecutionPolicy ByPass -c "irm https://chatgpt.com/codex/install.ps1 | iex"; alternatively use npm install -g @openai/codex or brew install --cask codex.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/codex.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.codex/skills/update-v8-version" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- update rusty_v8 artifacts,
- validate the V8 release candidate,
- diagnose why this V8 update no longer builds.
Ask for a task such as “bump V8,” “update rusty_v8 artifacts,” “validate the V8 release candidate,” “check v8-canary,” or “diagnose why this V8 update no longer builds.” The skill first reads third_party/v8/README.md, then performs the relevant updates and validation. The source does not define a separate CLI entry point or parameter format.
What are this skill's strengths and limitations?
- Covers the Cargo, Bazel, documentation, and checksum locations that carry the version pin.
- Includes checksum update and verification helpers plus a focused Python test suite.
- Prefers hosted v8-canary and distinguishes it from local substitute validation.
- Focuses failure investigation on build-relevant upstream deltas.
- Tightly coupled to Codex-specific paths, files, and workflows.
- Requires shell, filesystem access, and Python 3; hosted canary checks may also require network access and gh.
- The source provides no cross-platform test results or concrete success record.
- It does not publish tags, releases, or pushes.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| V8 Version Maintainer this page ✓ OpenAI · Official | 50 · Use with care | ★ 129k | 11d ago | Apache-2.0 |
| Azure Queue Storage for Rust ✓ Microsoft · Official | 54 · Use with care | ★ 2.7k | 3mo ago | MIT |
| Azure Blob Storage for Rust ✓ Microsoft · Official | 53 · Use with care | ★ 2.7k | 3mo ago | MIT |
| CPU Profile Analyzer ✓ Microsoft · Official | 52 · Use with care | ★ 194k | 11d ago | MIT |
| Remote Executor Integration Tests ✓ OpenAI · Official | 50 · Use with care | ★ 129k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The workflow is scoped to explicit V8 version, build, and CI surfaces, keeps checksum helpers involved, and explicitly forbids publishing, pushing, or tagging unless requested. It lacks pre-edit confirmation, a rollback procedure, and sensitive-data handling guidance, so points are deducted.
The sequence, happy path, failure path, and first actionable error reporting are reasonably clear, with concrete helper checks and unit-test commands. However, it depends on the README, GitHub/gh, tool availability, and environment conditions, and provides no static failure examples, so the score is limited by the static cap and uncertainty.
The name, trigger description, and V8 bump, canary, and artifact-build scenarios are clear, and a local substitute is allowed when hosted CI is unavailable. Non-fit boundaries, input/output contracts, and exclusion conditions are not explicit; Chinese support is absent, and GitHub-dependent operations may be difficult to reach from mainland China, so points are deducted.
The document is layered into core workflow, failure path, and reporting, with concrete files, commands, and agent interface metadata. The repository supplies Apache-2.0 licensing and verified OpenAI provenance, but the skill lacks its own version, changelog, maintenance ownership, installation prerequisites, FAQ, and fuller limitation disclosure, so points are deducted.
It can guide updates to multiple V8 pins, run focused validation, inspect canary status, and trace upstream build deltas, covering the core task. There are no committed real outputs, representative results, or alternative comparisons; under static calibration the score cannot exceed 7, so points are deducted.
Specific repository files, scripts, tests, and CI workflows are named, and the skill requires distinguishing hosted canary results from local substitutes, giving useful auditability. There are no execution records, test results, or independent corroboration, and static calibration caps the score at 5, so points are deducted.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →