agilab-local-llm

Guidance for using local LLM backends (Ollama/GPT-OSS) inside AGILAB with correctness-first prompts.

📖 中文导读:Guidance for using local LLM backends (Ollama/GPT-OSS) inside AGILAB with correctness-first prompts. 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。

name: agilab-local-llm description: Guidance for using local LLM backends (Ollama/GPT-OSS) inside AGILAB with correctness-first prompts. license: BSD-3-Clause (see repo LICENSE) metadata: updated: 2026-01-09

Local LLM Skill (AGILAB)

Use this skill when working on local/offline engines or prompts.

Correctness-First Defaults

  • Prefer deterministic settings for code edits (lower temperature, explicit constraints).
  • Require the model to return:
    • file list to edit
    • exact patch intent
    • tests/commands to validate

No Silent Fallbacks

  • Detect missing local endpoints/models up-front and surface an actionable error.
  • Do not auto-switch APIs or rewrite parameters silently.

Ollama Notes

  • Let users select:
    • model name
    • temperature/top_p/top_k
    • max tokens
    • seed (if supported)
    • “auto-run/auto-fix” loop guardrails (max iterations, stop on failure)