binrogithub

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Batch unit-test generation on the GLM execution pool (W2 fan-out). Use for "add tests for these N modules/files".

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Deploy Huawei CodeArts CLI on macOS and add a codearts-litellm wrapper that connects CodeArts to an OpenAI-compatible LiteLLM gateway, including model config, local auth proxy, language behavior patches, and session-scoped --yolo support. Use when installing CodeArts from scratch, configuring CodeArts with LiteLLM or…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Configure a side-by-side codex-glm command that routes Codex CLI to Huawei Cloud MaaS glm-5.1 through a CCR /v1/responses shim while preserving the original codex command.

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Install, configure, verify, or troubleshoot a side-by-side codex-forky command that routes normal Codex tool/code-execution turns through an existing external forky service to the MaaS execution backend such as LiteLLM/Huawei MaaS GLM, while routing non-tool ordinary turns and image turns directly to Codex…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Install, verify, diagnose, and harden GitHub Copilot Chat in VS Code when Huawei Cloud MaaS GLM or another Huawei MaaS OpenAI-compatible model is used through OAI Compatible Copilot. Use when Copilot stalls, emits malformed or truncated tool calls, writes partial files, shows path drift, needs Huawei MaaS model…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

USER-GLOBAL hybrid routing for Cursor (all workspaces): after install, writes /.cursor/rules + /.cursor/memory + /.cursor/hooks (sessionStart + beforeSubmitPrompt). Code execution (HTML/CSS/JS, Hello World, greenfield, 输出代码, tests, docs, CI) MUST use delegate.py → Huawei MaaS GLM — user need not name this skill. Plan…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Deploy, configure, verify, upgrade, troubleshoot, or remove a LiteLLM bridge for native Claude Code and OpenCode backed by Huawei MaaS GLM-5.2 and OpenRouter. Use for Anthropic Messages compatibility, GLM reasoning filtering, structured tool calls, multilingual smart routing, context length-band policy…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Deploy, verify, upgrade, or roll back the anthropicstreamguard LiteLLM plugin on a docker-compose LiteLLM gateway. Use when asked to install/uninstall the plugin, wire a LiteLLM deployment for native Claude Code clients, or verify the gateway stream fix.

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Task-level hybrid routing for OpenCode. Premium model GLM-5.2 handles architecture, complex debugging, security, incidents, high-risk review, images, and raw context over 128K. Execution tasks such as unit tests, docs, CI fixes, codegen, batch refactors, format or migration transforms, and low/medium-risk review are…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Benchmark and reduce LLM inference token cost on Huawei Ascend 910 NPUs with vLLM-Ascend and a Mooncake DRAM KV tier. Use when working on Qwen3.6-35B-A3B serving, 64K-context agent workloads, prefix-cache tiering, data-parallel session affinity, concurrency tuning, or when a throughput/cost benchmark gives results…

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Deploy and benchmark DeepSeek-V4-Flash-0731 W8A8 on Ascend A3 to achieve C64 aggregate output TPS >1280 (per-concurrency >20 tok/s). Use when the user asks to optimize DSV4 C64 throughput, enable Fused MC2, tune DSpark for high concurrency, or reproduce the 1523 tok/s result.

binrogithub/1-3-Cloud-Adoption-Skills

Skill Claude CodeCodex

Deploy and benchmark DeepSeek-V4-Flash-0731 W8A8 with DSpark speculative decoding on Ascend A3 NPU using vLLM 0.25.1. Use when the user asks to deploy DSV4 with DSpark, benchmark DSV4 on Ascend, enable speculative decoding for DeepSeek V4, or reproduce the DSpark TPS optimization test.