بنقرة واحدة
burrete-code-review
Run a final multi-angle code review on a Burrete branch or pull request.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Run a final multi-angle code review on a Burrete branch or pull request.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Preview supported molecular attachments or an explicit public PDB ID in Burrete's interactive viewer, or use the bounded hosted Ketcher editor for transient chemical sketches.
Use when an external agent must open, observe, control, or render into a Burrete molecular workspace without managing transport details.
Primary router for Burrete. Use when the plugin is at-mentioned or for molecular workspace work: opening structures, operating Mol* or Ketcher, reviewing SDF/property collections, trajectories, workflow result bundles, molecular reports, or Browser/Computer visual QA.
Use when opening molecular artifacts in Burrete Browser preview or desktop app sessions and establishing an observable workspace.
Use when visually verifying Burrete Browser preview or desktop app state without replacing typed observe/action contracts.
Use when molecular notes, property tables, charts, workflow reports, or provenance should be rendered beside an active Burrete workspace.
| name | burrete-code-review |
| description | Run a final multi-angle code review on a Burrete branch or pull request. |
Use this skill when asked to review a Burrete branch, PR, or large local change.
Run independent review passes for each focused review skill in this directory:
burrete-review-breaking-contractsburrete-review-change-sizeburrete-review-context-boundsburrete-review-testingburrete-review-path-contractsburrete-release-readiness when the diff touches release, installer,
versioning, signing, update, package manager, or packaging filesSubagents may be used: one subagent per focused review skill is preferred for large diffs. Pass the full skill path to each subagent and ask it to return only concrete findings.
Return findings first, ordered by severity. Every finding must include:
If there are no findings, say that clearly and list any residual risk or validation that was not run.