Automate pull request reviews with OpenCode CLI and GPT-5.4.
npx skills add https://github.com/dyoshikawa/webseek --skill opencode-gpt-review
Please help me install this Agent Skill. Skill: opencode-gpt-review Source: https://github.com/dyoshikawa/webseek/tree/main/.rulesync/skills/opencode-gpt-review Command: npx skills add https://github.com/dyoshikawa/webseek --skill opencode-gpt-review
This skill addresses the noise and hallucinations often found in automated AI code reviews by performing a multi-stage verification process that filters out invalid findings.
Run the opencode-gpt-review skill on the current branch to generate a vetted list of pull request improvements.
High-intent search queries and answers about installing and using this skill.
Automated pull request reviews reduce AI hallucinations through a multi-stage verification process that cross-references GPT-5.4 findings against the actual codebase. This contextual filtering distinguishes valid architectural issues from noise, presenting only actionable feedback to developers.
To run a GPT code review on your current branch, execute the opencode-gpt-review skill. It leverages the github-copilot/gpt-5.4 model via OpenCode to perform deep analysis and automatically generates a vetted list of pull request improvements.
Yes, you need active GitHub Copilot authentication to use OpenCode for code analysis. The automated review process requires this authentication alongside a local OpenCode installation to perform high-fidelity code inspection and validation.
Automated code review verifies AI findings by cross-referencing them against the actual codebase. This verification logic ensures accuracy and relevance, distinguishing valid architectural issues from AI hallucinations before presenting actionable feedback to the developer.
Using GPT-5.4 for pull request reviews minimizes false positives through contextual filtering. The skill automatically validates findings against the actual codebase to ensure accuracy, filtering out hallucinations and presenting only relevant architectural improvements.
AI code reviews generate irrelevant architectural suggestions due to hallucinations. This skill solves that problem by performing a multi-stage verification process that cross-references findings against the actual codebase to filter out invalid findings and ensure accuracy.
Moliboy5000Delegate code review and analysis tasks to GitHub Copilot via opencode CLI.
hyperlistEnforce Git branching, commit, and pull request standards with automated staging.
mkikenGenerate GitHub Pull Request titles and bodies from branch diffs.
BoraPerusicAutomate expert code review for multi-language codebases and pull requests.
