skills.rest
warpdotdev
warpdotdevOFFICIAL
Explore all skills by this author →
WebsiteGitHub

cross-platform-cloud-verification

Orchestrates cost-conscious verification of code changes across cloud OS and architecture runners.

64.7k|5.5k|Updated Jul 8, 2021
Search Tags:#Testing#QA automation#ci#Cross-Platform#runner orchestration#multi-os#cloud verification
One-click install
npx skills add https://github.com/warpdotdev/warp --skill cross-platform-cloud-verification
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cross-platform-cloud-verification
Source: https://github.com/warpdotdev/warp/tree/main/.agents/skills/cross-platform-cloud-verification
Command: npx skills add https://github.com/warpdotdev/warp --skill cross-platform-cloud-verification

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Verifying that a code change works across operating systems and CPU architectures is expensive and error-prone when done ad hoc. This Skill provides a disciplined workflow that runs cheap local checks first, selects the smallest relevant set of cloud runners, and aggregates results without hiding platform gaps.

Core Features & Use Cases

  • Runner Discovery and Selection: Discovers available runners at execution time via the Oz CLI and selects a minimal OS/architecture matrix based on evidence in the change, not runner availability.
  • Child Agent Orchestration: Launches verification-only child agents with run_agents, routing each to a specific runner and collecting structured pass/fail/blocked results with evidence.
  • Cost Guardrails: Enforces a local verification gate before cloud fan-out, limits retries for transient failures, and prevents children from fixing or pushing code.
  • Use Case: After fixing a filesystem path bug that affects Windows and macOS, use this Skill to verify the fix on one representative runner per affected OS, then receive a consolidated report showing which platforms passed and which remain unverified.

Quick Start

Verify my latest commit across the affected operating systems using the smallest set of cloud runners and report the results.

Frequently Asked Questions about cross-platform-cloud-verification

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify a code change across multiple operating systems in the cloud?▼

Run local builds and tests first, then discover runners with oz-dev runner list and select one representative runner per affected OS. Launch verification-only child agents via run_agents with remote.runner_id set to each selected runner UID, then aggregate the results.

How do I choose which platforms to test for a code change?▼

Infer affected dimensions from the diff: treat changes touching OS-gated code, packaging, filesystems, or shells as OS-sensitive, and only add architectures when there is evidence like FFI, SIMD, or architecture-specific bugs. Select one representative runner per affected OS.

When should I add multiple CPU architectures to the test matrix?▼

Add another architecture only when the change is architecture-sensitive, such as when it involves assembly, unsafe code, ABI concerns, endianness, native libraries, or an architecture-specific bug report. Do not add architectures merely because runners exist.

What happens if no runner is available for a required platform?▼

Missing runners do not block verification on available platforms. The final report marks the overall result as incomplete and explicitly lists each relevant OS or architecture that could not be verified due to lack of a suitable runner.

Can cloud verification replace local testing?▼

No. Cloud verification is the last layer after local builds, focused tests, lint, and type checks pass. Remote runs consume compute credits, so deterministic local failures must be fixed before launching any cloud fan-out.

What are the limitations of cross-platform cloud verification?▼

It requires the Oz CLI and run_agents support for remote.runner_id, consumes remote compute credits, and cannot verify platforms lacking available runners. Children are prohibited from fixing code, so failures must return to the parent workflow for local fixes.

Related Skills

Explore More Software Engineering Skills→
DxTaDxTa

push-all

Stage, check, commit, and push local Git changes with user confirmation.

Community
Advanced
mvillmowmvillmow
18

fix-ci-failures

Diagnose CI failures by inspecting logs and reproducing issues locally.

Community
Advanced
nebiusnebius
17

testing-conventions

Standardize testing protocols and infrastructure validation for the Nebius Physical AI repository.

Official
Advanced
minicooheiminicoohei
345

gcp-auth

Authenticate Google Cloud Platform using Application Default Credentials.

Community
Advanced