aiml-llm-reasoning
aimlapihello 作
Productivity & Tasks概要
Run AIMLAPI LLM and reasoning workflows through chat completions with retries, structured outputs, and explicit.
説明
このスキルでできること
Calls AIMLAPI chat completions via a Python script with automatic retries, optional API key file fallback, and a custom User-Agent header on every request. Supports reasoning parameters, temperature control, and structured JSON output. Results can be written directly to a file.
Built-in retry logic and API key file fallback reduce boilerplate compared to writing raw AIMLAPI requests by hand.
使いどころ
- Draft a project checklist from a plain-text prompt
- Generate a structured JSON risk list for a project
- Plan a feature rollout with step-by-step reasoning
- Summarize a long document into bullet points
- Script batch prompting runs against AIMLAPI models
ワークフロー例
User asks: Return a JSON array of 3 project risks with mitigation strategies
- 1
Builds request payload with json_object response_format and the user prompt
- 2
Sends POST to AIMLAPI chat completions endpoint with User-Agent header
- 3
Retries automatically on transient failures
- 4
Parses the response and writes JSON result to ./out/risks.json
A risks.json file containing a JSON array of 3 project risks each with a mitigation strategy
要件
- AIMLAPI API key set as the AIMLAPI_API_KEY environment variable
インストール
clawhub install aimlapihello/aiml-llm-reasoningCLIコマンドを実行するか、Claude Code、Cursor、Codexなどお使いのコーディングエージェントにプロンプトを貼り付けてください。