agent-improve
Use when evaluating agent skills or plugins, comparing versions, or running evidence-based improvement experiments, including changes to the improvement method itself. Not for ordinary task execution, routine skill authoring, one-off edits, or general design reviews.
Pinned to revision bac68dd4b8de, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/agent-improve/SKILL.md
- skills/agent-improve/agents/openai.yaml
- skills/agent-improve/references/experiments.md
- skills/agent-improve/references/host-execution.md
- skills/agent-improve/references/improvement.md
- skills/agent-improve/references/measurement.md
- skills/agent-improve/references/runtime.md
- skills/agent-improve/scripts/agent-eval.mjs
- skills/agent-improve/scripts/agent-eval.test.mjs
- skills/agent-improve/scripts/codex-adapter.mjs
- skills/agent-improve/scripts/comparison.mjs
- skills/agent-improve/scripts/comparison.test.mjs
- skills/agent-improve/scripts/experiment.mjs
- skills/agent-improve/scripts/host-adapter.mjs
- skills/agent-improve/scripts/records.mjs
Every link opens the file at its source, pinned to the revision this page describes.