feat(skills): NVIDIA/skills trusted tap + skills.sh.json categorization (re-open, awaiting NVIDIA) - #34817
Merged
teknium1 merged 2 commits intoMay 31, 2026
Conversation
NVIDIA/skills is now a default trusted tap in the Hermes Skills Hub — discoverable, browsable, searchable, and auto-updating through the same pipeline that already serves OpenAI, Anthropic, and HuggingFace skills. Rebased onto current main.
A GitHub tap can ship a repo-root skills.sh.json (the published skills.sh
schema) declaring category groupings. The Skills Hub now reads it at index
time and uses each grouping title as the skill's category label, instead of
the tag-derived guess. Generic: any tap that ships the file gets real
categorization — NVIDIA's groupings (Inference AI, Decision Optimization,
GPU Development, etc.) flow through automatically.
- GitHubSource: _get_skillsh_groupings() fetches+caches the sidecar per repo;
_parse_skillsh_groupings() flattens it to {skill_name: title};
_list_skills_in_repo() stamps meta.extra['category']; _meta_to_dict now
serializes extra so the category survives the index cache round-trip.
- extract-skills.py: prefers extra['category'] over the tag heuristic and
exempts sidecar categories from the small-category to Other collapse.
- Docs + 12 tests.
alt-glitch
added
type/feature
teknium1
merged commit May 31, 2026
fe141a2
into
revert-34333-feat/nvidia-skills-tap
8 checks passed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Re-opens the NVIDIA/skills tap integration +
skills.sh.jsoncategorization that was merged in #34333 and reverted in #34815. This PR is intentionally parked for review — do not merge until NVIDIA confirms (see Coordination).NVIDIA/skillsbecomes a default trusted tap in the Hermes Skills Hub — discoverable, browsable, searchable, auto-updating through the same pipeline that serves OpenAI, Anthropic, and HuggingFace skills. On top of that, any GitHub tap that ships a repo-rootskills.sh.jsongrouping sidecar gets real category labels in the Skills Hub instead of a tag-derived guess.Changes
Tap integration
tools/skills_hub.py—NVIDIA/skillsadded toGitHubSource.DEFAULT_TAPS(path:skills/).tools/skills_guard.py—NVIDIA/skillsadded toTRUSTED_REPOS(trust_level="trusted").website/scripts/extract-skills.py—NVIDIA/skills→"NVIDIA"inGITHUB_TAP_LABELS.website/src/pages/skills/index.tsx—NVIDIApill inSOURCE_CONFIG(green#76b900), slotted after HuggingFace.website/docs/user-guide/features/skills.md(+zh-Hans) — NVIDIA added to default-taps and trusted-repos lists.Category groupings via
skills.sh.json(generic, not NVIDIA-specific)tools/skills_hub.py—GitHubSource._get_skillsh_groupings()fetches+caches the repo-rootskills.sh.jsonper repo;_parse_skillsh_groupings()flattens itsgroupings(title+skills[]) into{skill_name: title};_list_skills_in_repo()stampsSkillMeta.extra["category"];_meta_to_dict()serializesextraso the category survives the index cache round-trip.website/scripts/extract-skills.py— prefersextra["category"]over the tag heuristic; exempts sidecar categories from the small-category → "Other" collapse.website/docs/user-guide/features/skills.md— documents the mechanism with an example.We standardized on the published
skills.sh.jsonschema rather than NVIDIA's proprietarymetadata.json, so any future tap adopting the standard gets categorization for free.Validation
tests/tools/test_skills_hub.py + test_skills_guard.pyskills.sh.json_get_skillsh_groupings(caching + no-sidecar→None)extract-skills.pyconsolidation (sidecar category survives; heuristic small-cats still collapse)Coordination — DO NOT MERGE YET
Awaiting NVIDIA confirmation (@mhollinger / @moshe / @nikhilkswami) that
skills.sh.jsonis the sidecar they want us standardized on. NVIDIA's open PR (NVIDIA/skills#90) ships two sidecars (metadata.json+skills.sh.json); this integration readsskills.sh.json. Their stated launch plan is domain-only filtering, which maps cleanly onto theskills.sh.jsongroupings. Hold the merge until they comment.