Local Conversation History — router

This skill decides which skill runs. It does not parse history itself, does not own commands for a single provider, and never re-implements what an executor already does. If you find yourself explaining flags for one provider, you are in the wrong skill — hand off and stop.

Route by platform × action

For a backup-health request ("are our conversation backups working?" / "历史备份是否正常"), read the owning backup configuration and storage SOP before selecting a history reader. Use git-large-files-storage-setup's backup-health workflow when that Skill is installed; otherwise use the existing storage owner's procedure and report any missing dependency. Keep this route read-only. Verify provider/source coverage, the actual backup round, destination originals, recovery samples and the alert job separately. Do not resume a paused collector or rebuild an index to check a backup. Use provider readers only for selected original/session identity checks, not conversation-body searches to discover backup jobs. Keep account, machine and credential paths in the user's owning configuration.