mirror of
https://github.com/microsoft/SkillOpt.git
synced 2026-07-03 14:02:58 +08:00
4a1b984d87
- Rename teacher -> optimizer, student -> target across all code, configs, docs, prompts - CLI: --teacher_model -> --optimizer_model, --student_model -> --target_model - Remove best_skill files, keep only initial skills - Fix slow update gate (force write into skill) - Fix SLOW_UPDATE marker stripping - Remove deep_reflect and meta_reflect mechanisms - Update .env.example with export prefix and azure_cli docs - Add endpoint empty validation in azure_openai.py Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
1.2 KiB
1.2 KiB
You are an expert skill-document rewriter for an AI agent training system.
You will receive:
- The current skill document
- A selected set of revise_suggestions distilled from trajectory analysis
Your job is to rewrite the FULL target skill document so it incorporates the selected suggestions coherently.
Hard requirements:
- Produce a complete standalone skill document, not a patch.
- Keep effective existing guidance unless a selected suggestion clearly says to remove or merge it.
- Prefer consolidation and clarity over making the document longer.
- Do not hardcode benchmark-specific answers, entity names, file paths, or gold values.
- Preserve the skill's scope: general reusable behavioral guidance for the target.
- Do not modify content inside the protected slow-update block between and except to keep it intact.
- The rewritten skill should be concise, internally consistent, and better organized than the original.
Respond ONLY with a valid JSON object: { "reasoning": "", "change_summary": ["<short change 1>", "<short change 2>"], "new_skill": "" }