Improvement Orchestrator

Coordinates the full improvement pipeline: Generator → Discriminator → Evaluator → Executor → Gate.

When to Use

• Run a full improvement cycle on one or more skills
• Coordinate the 5-stage pipeline end-to-end (with optional evaluator)
• Retry failed improvements with trace-aware feedback (Ralph Wiggum loop)

When NOT to Use

• 只想检查 skill 质量评分 → use improvement-learner
• 只想手动给候选打分 → use improvement-discriminator
• 只想改一个文件 → use improvement-executor
• 只想查基准数据 → use benchmark-store

Pipeline

propose → discriminate → evaluate* → execute → gate
         ↻ Ralph Wiggum: fail → inject trace → retry (max 3)
         * evaluate is optional — skipped if no task_suite.yaml exists

\x3Cexample> 正确用法: 对一个 skill 运行全流程改进 $ python3 scripts/orchestrate.py --target /path/to/skill --state-root ./state → 自动完成: 生成候选 → 多人盲审 → 任务评估 → 执行变更 → 6层门禁 → 失败时自动注入 trace 重试（最多 3 次） \x3C/example>

\x3Canti-example> 错误用法: 只想看评分却用了 orchestrator $ python3 scripts/orchestrate.py --target /path/to/skill # 会执行变更！ → 应该用: python3 improvement-learner/scripts/self_improve.py --skill-path /path/to/skill --max-iterations 1 \x3C/anti-example>

CLI

python3 scripts/orchestrate.py \
  --target /path/to/skill \
  --state-root /path/to/state \
  --max-retries 3 \
  --auto

Output Artifacts

Related Skills

• improvement-generator: Produces candidate proposals (stage 1)
• improvement-discriminator: Multi-reviewer panel scoring (stage 2)
• improvement-evaluator: Task suite execution validation (stage 3, optional)
• improvement-executor: Applies changes with backup/rollback (stage 4)
• improvement-gate: 6-layer quality gate (stage 5)
• benchmark-store: Frozen benchmarks and Pareto front data

References

• Architecture — System design and data flow
• Guardrails — Safety rules and protected targets
• End-to-End Demo — Complete walkthrough