components.d/tilegym: flip source path from .agents/skills/ to skills/#121
Merged
Merged
Conversation
TileGym PR NVIDIA/TileGym#132 migrated all 7 cuTile skill folders from `.agents/skills/` to the canonical release-facing `skills/` path. The previous `.agents/skills` was retained as a backward-compatibility symlink for tools that still expect the agentskills.io layout, but the sync workflow's git sparse-checkout + rsync does not reliably follow that symlink — it would either copy just the symlink-as-file or miss the contents. Update `path` from `.agents/skills/` → `skills/` so the sync workflow reads from the public-facing path directly. Single-entry parent-path form is retained (rather than per-skill flat layout) because all 7 cuTile skills are intended for the public catalog — there is no contributor/maintainer-only skill subset to filter out — and the parent-path form auto-discovers any new TileGym skills without requiring further nvidia/skills PRs. Also add the standard Apache SPDX header to match the convention adopted across components.d/ on 2026-05-28 (cf. NVIDIA#109, NVIDIA#113). Catalog effect: skills under `nvidia/skills/skills/TileGym/<name>/` will continue to populate via the existing per-skill sync-compliance gate (`skill.oms.sig` + `skill-card.md` + `evals.json`). The first TileGym skills landing `evals.json` are tracked in NVIDIA/TileGym#135 (adding-cutile-kernel, cutile-python); the remaining 5 will follow in subsequent PRs. Signed-off-by: Hannah Li <hanli@nvidia.com>
hannahli-nv
added a commit
to NVIDIA/TileGym
that referenced
this pull request
May 29, 2026
Adds the skill evaluation dataset for `adding-cutile-kernel`. The question targets two TileGym CI naming conventions required for new operators — the `test_op*` test-function prefix that gates which pytest functions are collected by the CI `-k test_op` filter, and the `-TFLOPS` / `-GBps` suffix required for the benchmark `plot_name` parameter so results are parsed into the CI summary. These conventions are documented in the adding-cutile-kernel `SKILL.md` and are not reliably available from general training data, so the skill provides clear lift over the no-skill baseline. The case has produced zero high-severity regression findings across three prior nvskills-ci runs. Schema fields used: `id`, `question`, `expected_skill`, `expected_script`, `ground_truth`, `expected_behavior`. After this PR merges, the publication pipeline auto-generates `BENCHMARK.md`, `skill-card.md`, and the detached signature `skill.oms.sig` for this skill, and the nvidia/skills sync workflow publishes it to the public catalog (per NVIDIA/skills#121). The remaining 6 cuTile skills will receive their own `evals/evals.json` in follow-up PRs, scoped per-skill to keep each evaluation run within the per-job time budget and avoid blocking each other through the global gate. Signed-off-by: Hannah Li <hanli@nvidia.com>
4 tasks
mosheabr
approved these changes
May 29, 2026
Collaborator
mosheabr
left a comment
mosheabr
left a comment
There was a problem hiding this comment.
Approved. Hannah's source-side work is complete: all 7 cuTile skills renamed with tilegym- prefix and migrated to canonical skills/ path. 1 of 7 (tilegym-adding-cutile-kernel) is fully 5/5; the other 6 still need eval datasets added before they'll pass sync enforcement. Merging this legacy-sweep entry now to land the path correction; will follow up immediately with a flat-layout restructure PR (no sync runs between merges).
mosheabr
added a commit
that referenced
this pull request
May 29, 2026
Follow-up to PR #121 (Hannah's path flip from .agents/skills/ to canonical skills/). Restructure to one-entry-per-skill flat layout matching the tilegym- prefixed names Hannah landed in source. PR #121 was merged with the legacy sweep pattern (path: skills/, catalog_dir: TileGym) to land the canonical-path correction quickly. Merging this PR before the first post-#121 sync runs means no nested skills/TileGym/ directory is ever created in the catalog. Source state (7 skills, all sig + skill-card + BENCHMARK.md): | Skill | sig | card | evals | BENCHMARK | |---|---|---|---|---| | tilegym-adding-cutile-kernel | ✓ | ✓ | ✓ | ✓ | | tilegym-converting-cutile-to-julia | ✓ | ✓ | ✗ | ✓ | | tilegym-converting-cutile-to-triton | ✓ | ✓ | ✗ | ✓ | | tilegym-cutile-autotuning | ✓ | ✓ | ✗ | ✓ | | tilegym-cutile-python | ✓ | ✓ | ✗ | ✓ | | tilegym-improve-cutile-kernel-perf | ✓ | ✓ | ✗ | ✓ | | tilegym-monkey-patch-kernels-to-transformers | ✓ | ✓ | ✗ | ✓ | Sync after this lands publishes tilegym-adding-cutile-kernel flat at top level. The other 6 will drop on enforcement until Hannah's team adds eval datasets — Slack drafted to track the gap. Signed-off-by: Moshe Abramovitch <moshea@nvidia.com>
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
Flip
components.d/tilegym.yml'spathfrom.agents/skills/→skills/.TileGym NVIDIA/TileGym#132 migrates all 7 cuTile skill folders from
.agents/skills/to the canonical release-facingskills/path. The previous.agents/skills/is retained as a backward-compatibility symlink for tools that still expect the agentskills.io layout, but the sync workflow'sgit sparse-checkout+rsyncdoes not reliably follow that symlink — it would either copy just the symlink-as-file or miss the contents entirely.TileGym NVIDIA/TileGym#135 (now merged) onboarded all 7 cuTile skills to
nvskills-ci: skills moved intoskills/, each folder got atilegym-prefix, NVSkills validation signatures +skill-card.md+BENCHMARK.mdare attached for every skill, andtilegym-adding-cutile-kernelships with a 5-caseevals/evals.json(Tier 3 verdict: PASS, codex overall lift+0.02, best-performing agent codex).Single-entry parent-path form (
path: skills/,catalog_dir: TileGym) is retained rather than per-skill flat layout because:evals.jsonto additional skills in follow-up TileGym PRs (6 more cuTile skills are queued) will not require additionalnvidia/skillsPRs.Also adds the standard Apache SPDX header to match the convention adopted across
components.d/on 2026-05-28 (cf. #109, #113).Source state — compliance per skill (after TileGym #135 merged)
The 1 fully compliant skill (
tilegym-adding-cutile-kernel) syncs tonvidia/skills/skills/TileGym/tilegym-adding-cutile-kernel/when this PR lands. The remaining 6 stay dropped by the sync compliance gate until theirevals/evals.jsonlands in subsequent TileGym PRs (these will sync automatically without further changes here).Test plan
skills/TileGym/tilegym-adding-cutile-kernel/appears withskill.oms.sig,skill-card.md,evals.json, andBENCHMARK.md.