القائمة
Run benchmark suites and manage policy evolution — create challengers, compare against champions, promote or rollback policies.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | benchmark |
| description | Run benchmark suites and manage policy evolution — create challengers, compare against champions, promote or rollback policies. |
Use this skill to benchmark policies and drive champion/challenger evolution.
autodialectics-mcp must be on PATH (pip install autodialectics).
benchmark(suite_dir?, policy_id?) — run the benchmark suite against a policy. Returns case-by-case results with scores and decisions.evolve_policy(use_gepa?) — analyze recent benchmark reports and create a challenger policy. Set use_gepa: false to skip the GEPA optimizer (simpler heuristic fallback).promote_policy(policy_id) — promote a challenger to champion if comparison rules allow.rollback_policy() — revert to the previous champion if the current one regresses.autodialectics benchmark
autodialectics evolve
autodialectics promote <policy_id>
autodialectics rollback
benchmark → evolve_policy → benchmark (with challenger) → compare → promote or rollback
evolve_policy returns no_reports, run benchmarks first to generate data.If the user passes a suite directory after /autodialectics:benchmark, use it as the benchmark suite path.
Examine Autodialectics run results, manifests, and stored artifacts. Use after a pipeline run completes or to review past runs.
Re-run a stored Autodialectics pipeline run, optionally with a different policy, to compare outcomes or debug regressions.
Compile and execute an Autodialectics anti-slop pipeline for a task. Covers health checks, runtime init, contract compilation, and full pipeline execution.
Use the local Autodialectics MCP server and CLI to compile tasks, execute anti-slop runs, inspect artifacts, replay runs, benchmark policies, and evolve champions in this repository.