Orchestrate multi-backend AI code tasks with pluggable backends and file references.
npx skills add https://github.com/cexll/myclaude --skill codeagent
Please help me install this Agent Skill. Skill: codeagent Source: https://github.com/cexll/myclaude/tree/main/skills/codeagent Command: npx skills add https://github.com/cexll/myclaude --skill codeagent
This Skill provides a pluggable, multi-backend wrapper for AI-assisted code tasks. It enables backends such as Codex, Claude, and Gemini to be selected per task, supports file references using @ syntax, and returns structured outputs to streamline automated coding workflows.
Use a simple task to validate integration: codeagent-wrapper "analyze @src/utils" [working_dir]
High-intent search queries and answers about installing and using this skill.
Multi-backend code task orchestration lets you select different AI backends—Codex, Claude, or Gemini—per task to analyze, refactor, or generate code. Use codeagent-wrapper with your chosen backend and file references using @ syntax to execute tasks across your codebase efficiently.
Yes, file references using @ syntax let you drive code analysis and generation across multiple project files in a single task. This eliminates manual file handling and keeps your workflow streamlined when working with large codebases.
Parallel execution orchestration lets you run code generation and analysis tasks concurrently across multiple backends and working directories. This reduces task completion time while maintaining security controls and per-task backend selection for optimal results.
Per-task backend switching means you can override your default backend for individual tasks without reconfiguring your entire setup. This flexibility lets you select the most suitable backend—Codex, Claude, or Gemini—based on each task's requirements.
Security controls are built into the multi-backend wrapper, allowing you to manage permissions and safety checks per backend. Combined with environment-variable configurability, this ensures secure execution across all your code analysis and generation workflows.
Yes, structured output formats from codeagent-wrapper enable seamless integration with automation pipelines. The standardized results from code analysis, generation, and refactoring tasks work directly with downstream CI/CD systems and reporting tools.
Snowflake-LabsA concise summary of your roleplay character: name,</name> personality, and/or behavior, formatted as a single string sentence. Typically不超过50字，此处省略
tile-aiValidate Python and C++ code formatting against CI rules and generate JSON/Markdown reports.
CapplequoppeAutomate code commits with testing, staging checks, and conventional messages.
tense-iCoordinate multi-backend AI code tasks with parallel execution and per-task backend selection.
