Rank the slash menu by the skills you actually use - #73704
Merged
Conversation
The catalog advertises every skill command but nothing about which ones the user actually reaches for, so consumers can only sort them alphabetically. Add a `skills` map keyed by slash command carrying the activity count and origin (hub / bundled / local) already tracked in the skills sidecars, read once per catalog build. Additive: older clients ignore the field, and an unreadable sidecar degrades to zero usage rather than failing the catalog.
The `/` popover listed skills alphabetically, so on a 200-skill install the ones invoked daily sat below a wall of skills that shipped with Hermes and were never opened — /research-paper-writing (never used) outranked /research (60 invocations). Sort the Skills section most-used first, A-Z within a tie, and on a bare `/` drop bundled skills with no recorded activity. Typing a query keeps every match: a search that hides a result is broken, so a typed `/re` only reorders.
OutThisLife
enabled auto-merge
July 29, 2026 00:18
Contributor
૮ >ﻌ< ა ci reviewran on 959d299 ℹ️ InfoDesktop E2E visual evidence · View test artifacts · View job1 visual diff. inline evidence is publishing... |
alt-glitch
added
type/feature
randlee
pushed a commit
to randlee/hermes-agent
that referenced
this pull request
Aug 11, 2026
…ls-by-usage Rank the slash menu by the skills you actually use
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The
/menu listed skills alphabetically, which means the ordering carries no information about what the user actually does. On a 200-skill install that buries the handful of skills someone invokes daily under a wall of skills that shipped with Hermes and were never opened —/research-paper-writing(0 activity) sorted above/research(60 invocations).Hermes already tracks per-skill activity in
~/.hermes/skills/.usage.jsonand surfaces it on the Capabilities page; the slash menu just never saw it. This plumbs that number into the command catalog and ranks on it.Gateway.
commands.cataloggains askillsmap keyed by slash command, each entry carryingusage(use + view + patch, the same count Capabilities shows) andorigin(hub/bundled/local, matching/api/skillsprovenance). Both come from sidecar files read once per catalog build. The field is additive — older clients ignore it — and an unreadable sidecar degrades to zero usage rather than failing the catalog.Desktop. The Skills section sorts most-used first, A–Z within a tie. On a bare
/— browsing, not searching — bundled skills with no recorded activity are dropped: shipped-and-ignored is noise, but the user's own unused skills and anything the catalog hasn't classified stay. Typing a query only reorders and never hides, because a search that drops a match is broken. The usage map rides along on the catalog response the popover has already fetched, so ranking a typed query costs no extra round trip.Measured against a real 200-skill install: 125 skills used at least once, 75 never — 38 of which are bundled and now stay out of the browse list.