evaluate-behavior
Use when testing or comparing how an agent behaves, such as whether a skill triggers or a setup change changes outcomes. Not for testing code (Verify Work).
Pinned to revision 51b9682fb44a, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/evaluate-behavior/SKILL.md
- skills/evaluate-behavior/references/authoring.md
- skills/evaluate-behavior/references/coding-tasks.md
- skills/evaluate-behavior/references/conversations-pipelines-judges.md
- skills/evaluate-behavior/references/design-and-measurement.md
- skills/evaluate-behavior/references/running-and-evidence.md
- skills/evaluate-behavior/scripts/claude-guard.mjs
- skills/evaluate-behavior/scripts/guard-rules.mjs
- skills/evaluate-behavior/scripts/lib/agreement.mjs
- skills/evaluate-behavior/scripts/lib/arms.mjs
- skills/evaluate-behavior/scripts/lib/budget.mjs
- skills/evaluate-behavior/scripts/lib/build-source.mjs
- skills/evaluate-behavior/scripts/lib/clear.mjs
- skills/evaluate-behavior/scripts/lib/container.mjs
- skills/evaluate-behavior/scripts/lib/decision.mjs
- skills/evaluate-behavior/scripts/lib/definitions.mjs
- skills/evaluate-behavior/scripts/lib/evidence.mjs
- skills/evaluate-behavior/scripts/lib/freeflow.mjs
- skills/evaluate-behavior/scripts/lib/grade-command.mjs
- skills/evaluate-behavior/scripts/lib/grade-comparison.mjs
- skills/evaluate-behavior/scripts/lib/grade-interaction.mjs
- skills/evaluate-behavior/scripts/lib/grade-support.mjs
- skills/evaluate-behavior/scripts/lib/grade-workspace.mjs
- skills/evaluate-behavior/scripts/lib/grade.mjs
- skills/evaluate-behavior/scripts/lib/hooks.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/confinement.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/constraints.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/index.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/isolation.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/launch.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/model-call.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/recorder.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/session.mjs
- skills/evaluate-behavior/scripts/lib/hosts/claude-code/skills.mjs
- skills/evaluate-behavior/scripts/lib/hosts/index.mjs
- skills/evaluate-behavior/scripts/lib/hosts/names.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/confinement.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/host.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/index.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/isolation.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/model-call.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/process.mjs
- skills/evaluate-behavior/scripts/lib/hosts/pi/rpc.mjs
- skills/evaluate-behavior/scripts/lib/identity.mjs
- skills/evaluate-behavior/scripts/lib/judge.mjs
- skills/evaluate-behavior/scripts/lib/metrics.mjs
- skills/evaluate-behavior/scripts/lib/model-call.mjs
- skills/evaluate-behavior/scripts/lib/participants.mjs
- skills/evaluate-behavior/scripts/lib/process.mjs
- skills/evaluate-behavior/scripts/lib/queue.mjs
- skills/evaluate-behavior/scripts/lib/reference.mjs
- skills/evaluate-behavior/scripts/lib/reuse.mjs
- skills/evaluate-behavior/scripts/lib/review.mjs
- skills/evaluate-behavior/scripts/lib/runner.mjs
- skills/evaluate-behavior/scripts/lib/sandbox.mjs
- skills/evaluate-behavior/scripts/lib/scan-shell.mjs
- skills/evaluate-behavior/scripts/lib/scan.mjs
- skills/evaluate-behavior/scripts/lib/seal.mjs
- skills/evaluate-behavior/scripts/lib/shell.mjs
- skills/evaluate-behavior/scripts/lib/source-override.mjs
- skills/evaluate-behavior/scripts/lib/subject-file.mjs
- skills/evaluate-behavior/scripts/lib/subject.mjs
- skills/evaluate-behavior/scripts/lib/task-copy.mjs
- skills/evaluate-behavior/scripts/lib/trials.mjs
- skills/evaluate-behavior/scripts/lib/validity.mjs
- skills/evaluate-behavior/scripts/lib/view-html.mjs
- skills/evaluate-behavior/scripts/lib/view.mjs
- skills/evaluate-behavior/scripts/pi-command-tool.mjs
- skills/evaluate-behavior/scripts/pi-container-route.mjs
- skills/evaluate-behavior/scripts/pi-guard.mjs
- skills/evaluate-behavior/scripts/pi-login-refresh.mjs
- skills/evaluate-behavior/scripts/pi-observer.mjs
- skills/evaluate-behavior/scripts/skill-eval.mjs
Every link opens the file at its source, pinned to the revision this page describes.