Rich history search and metadata

[raiseCatError]

Goal

Native history that uses what NMSh already knows, with structured search.

NMSh already records much more than flat zsh history. Native history should eventually support:

• command
• cwd
• project
• exit status
• duration
• date/time
• session
• success/failure
• frecency
• possibly command-sequence context

UX / product direction

Structured filters in /history search, for example:

cwd:   project:   exit:   before:   after:   session:   duration:

• /history = command-level search. /resume = session-level continuity. This does not replace /resume.
• Selecting a result restores it into the editor without running it (current Bare /history should enter interactive history search #120 behavior).

Scope

• A local, private metadata store. Derive or index from the session journal (Continuous session journal and advanced /resume #88) where possible rather than duplicating it.
• Import existing zsh history so users start with their past commands.
• Privacy: leading-space and HISTORY_IGNORE rules; ability to delete entries.
• Responsive with 100k+ entries.

Non-goals

• Sync or cloud features.
• Replacing /resume.

Acceptance criteria

• Commands run in NMSh are recorded with the metadata above.
• Filters work and combine; plain text search keeps working.
• Search stays responsive on a large benchmark history.
• Ignored commands are never stored.

Dependencies / related

• Part of Provider framework and unified provider UX #134 (History/Search family). Child: Atuin history provider #144 Atuin.
• Feeds NMSh Native Suggestions v2 #138 ranking and Smart directory navigation and zoxide integration #147 directory frecency. Related Continuous session journal and advanced /resume #88, Bare /history should enter interactive history search #120.

Research references

• Atuin: https://github.com/atuinsh/atuin
• McFly: https://github.com/cantino/mcfly (reference only; no separate issue planned)

[raiseCatError]

Development implementation is reviewable in #235, with cumulative hardening in #240; final acceptance documentation is #241. These are dependent, unmerged PRs on current dev, with no release/version bump.

The complete code head passed 714 local tests, build, typecheck, benchmark types and diff checks, plus Node 22/26 CI. Architecture and limitations describe what is implemented and deferred. One physical-QA checklist covers the relevant keyboard/layout/provider/privacy checks; it has not been performed.

This issue stays open until release. Project status could not be updated because the available token lacks read:project; active branches/PRs provide durable work status.