MANDATORY: Follow the complete shared lifecycle contract in ${CLAUDE_PLUGIN_ROOT}/skills/shared/run-lifecycle.md with declared skill difficulty-calibration.

difficulty-calibration

Use $ARGUMENTS as optional CLI flags for the analyzer.

Verification

After running the backing script, validate that the output includes a difficulty classification and any required evidence before reporting the result.

MANDATORY: READ ENTIRE FILE before composing user-facing prose: ${CLAUDE_PLUGIN_ROOT}/skills/shared/readability-style.md.

Analyze predicted versus realized difficulty tiers from the synchronized run-log cache.

The analyzer is read-only unless --out FILE is provided. It syncs once, then reads only the unpacked cache. It does not change thresholds, panels, reviewer points, token allocation, or live routing.

Usage