Pass agent skill - #4950
Open
mitchdz wants to merge 28 commits into
Open
Conversation
Collaborator
Author
Command Bot: Processing... |
taalexander
force-pushed
the
pass-agent-skill
branch
from
July 20, 2026 17:56
9a49384 to
bee4016
Compare
CI Summary (
|
| Job | Result |
|---|---|
binaries |
⏩ skipped |
build_and_test |
✅ success |
config_devdeps |
✅ success |
config_source_build |
⏩ skipped |
config_wheeldeps |
✅ success |
devdeps |
✅ success |
docker_image |
⏩ skipped |
gen_code_coverage |
⏩ skipped |
metadata |
✅ success |
python_metapackages |
⏩ skipped |
python_wheels |
⏩ skipped |
source_build |
⏩ skipped |
wheeldeps |
✅ success |
⏩ Skipped jobs (7) — intentionally skipped on PR builds; run on merge_group / workflow_dispatch
| Job |
|---|
binaries |
config_source_build |
docker_image |
gen_code_coverage |
python_metapackages |
python_wheels |
source_build |
All sub-jobs (42) — every matrix leg, with links
| Job | Status | Link |
|---|---|---|
| Build and test (amd64, gcc12, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (amd64, gcc12, openmpi) / Dev environment (Python) | ✅ success | view |
| Build and test (amd64, llvm, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (amd64, llvm, openmpi) / Dev environment (Python) | ✅ success | view |
| Build and test (arm64, llvm, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (arm64, llvm, openmpi) / Dev environment (Python) | ✅ success | view |
| CI Summary | ❔ in_progress | view |
| Configure build (devdeps) | ✅ success | view |
| Configure build (source_build) | ⏩ skipped | view |
| Configure build (wheeldeps) | ✅ success | view |
| Create CUDA Quantum installer | ⏩ skipped | view |
| Create Docker images | ⏩ skipped | view |
| Create Python metapackages | ⏩ skipped | view |
| Create Python wheels | ⏩ skipped | view |
| Gen code coverage | ⏩ skipped | view |
| Load dependencies (amd64, gcc12) / Caching | ✅ success | view |
| Load dependencies (amd64, gcc12) / Finalize | ✅ success | view |
| Load dependencies (amd64, gcc12) / Metadata | ✅ success | view |
| Load dependencies (amd64, llvm) / Caching | ✅ success | view |
| Load dependencies (amd64, llvm) / Finalize | ✅ success | view |
| Load dependencies (amd64, llvm) / Metadata | ✅ success | view |
| Load dependencies (arm64, gcc12) / Caching | ✅ success | view |
| Load dependencies (arm64, gcc12) / Finalize | ✅ success | view |
| Load dependencies (arm64, gcc12) / Metadata | ✅ success | view |
| Load dependencies (arm64, llvm) / Caching | ✅ success | view |
| Load dependencies (arm64, llvm) / Finalize | ✅ success | view |
| Load dependencies (arm64, llvm) / Metadata | ✅ success | view |
| Load source build cache | ⏩ skipped | view |
| Load wheel dependencies (amd64, 12.6) / Caching | ✅ success | view |
| Load wheel dependencies (amd64, 12.6) / Finalize | ✅ success | view |
| Load wheel dependencies (amd64, 12.6) / Metadata | ✅ success | view |
| Load wheel dependencies (amd64, 13.0) / Caching | ✅ success | view |
| Load wheel dependencies (amd64, 13.0) / Finalize | ✅ success | view |
| Load wheel dependencies (amd64, 13.0) / Metadata | ✅ success | view |
| Load wheel dependencies (arm64, 12.6) / Caching | ✅ success | view |
| Load wheel dependencies (arm64, 12.6) / Finalize | ✅ success | view |
| Load wheel dependencies (arm64, 12.6) / Metadata | ✅ success | view |
| Load wheel dependencies (arm64, 13.0) / Caching | ✅ success | view |
| Load wheel dependencies (arm64, 13.0) / Finalize | ✅ success | view |
| Load wheel dependencies (arm64, 13.0) / Metadata | ✅ success | view |
| Prepare cache clean-up | ❔ in_progress | view |
| Retrieve PR info | ✅ success | view |
✅ Required checks (6/6) — declared in .github/required-checks.yml for push
| Required check | Status | Link |
|---|---|---|
| Build and test (amd64, llvm, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (amd64, llvm, openmpi) / Dev environment (Python) | ✅ success | view |
| Build and test (arm64, llvm, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (arm64, llvm, openmpi) / Dev environment (Python) | ✅ success | view |
| Build and test (amd64, gcc12, openmpi) / Dev environment (Debug) | ✅ success | view |
| Build and test (amd64, gcc12, openmpi) / Dev environment (Python) | ✅ success | view |
Collaborator
Author
Command Bot: Processing... |
2 similar comments
Collaborator
Author
Command Bot: Processing... |
Collaborator
Author
Command Bot: Processing... |
taalexander
force-pushed
the
pass-agent-skill
branch
2 times, most recently
from
July 21, 2026 01:24
df25c8f to
840e5b3
Compare
Collaborator
Author
Command Bot: Processing... |
taalexander
force-pushed
the
pass-agent-skill
branch
from
July 22, 2026 14:29
b5371a1 to
ae475a7
Compare
Collaborator
Author
Command Bot: Processing... |
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Keep complete dialect reference generation behind the dialect-specific CMake helper and clarify the CC and Quake descriptions based on reviewer feedback. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Keep generated dialect references out of the public documentation build without leaving unresolved document roles for Sphinx to diagnose under -W. Preserve a working aggregate reference link in opt-in compiler developer builds and pluralize the compiler IR page title per review. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
taalexander
force-pushed
the
pass-agent-skill
branch
3 times, most recently
from
July 23, 2026 19:17
23334e1 to
7c67ca3
Compare
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Add a source-editing evaluation that asks agents to build and test an out-of-tree H-X-H to Z pass without a supplied scaffold. Normalize the evaluator configuration to the current SkillEvaluator schema and document qualification, leakage, and grading requirements. The CUDA-Q development image could not be staged because its final 9.16 GB layer did not complete after two pulls. Record that infrastructure blocker without claiming agent or deterministic grader results, and keep the case out of pull request smoke pending sanitized staging and a validated private build grader. Verification: JSON and YAML parsing, SkillEvaluator config and dataset parsing, quality-check (83/B), markdown link check, git diff checks, repository pre-commit hooks, and independent read-only review. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Run the external H-X-H plugin case as an authoritative native Harbor task so build and correctness failures gate qualification. Stage the CUDA-Q development environment without exposing evaluator assets, and verify the agent project with CMake, CTest, cudaq-opt, FileCheck, and CircuitCheck under bounded CI resources. Validated the public native-task staging contract for skill and baseline conditions, reward failure handling, static formats, and repository pre-commit hooks. The container integration remains for NVSkills CI because the CUDA-Q development image is not available locally. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Remove redundant output guidance from the opening paragraph so the skill starts with its purpose and source routing. The broader completion guidance continues to prevent copied policy and source material. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Replace the vague change-shape term with compiler extension point in the skill and evaluation expectations. The section now states directly that contributors must choose among folding, canonicalization, analyses, interfaces, IR changes, an existing pass, or a focused new pass. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Evaluate the pass-development skill with SkillEvaluator's standard trajectory and answer grader in both conditions. Preserve a CUDA-Q development container so the external-plugin case can build and test against the warm compiler environment. Remove the native custom grader and hidden fixture, consolidate overlapping cases, and document the remaining trajectory-grading boundary. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Rebase the pass-development eval Dockerfile onto the SkillEvaluator base image and reuse that base so the evaluator can provision its agent harness. The preserve/disabled modes were left over from the removed native-harbor task; with them no agent was eligible, Tier 3 execution was skipped, and the NVSkills merge gate reported canonical Tier 3 coverage as invalid (reason code tier3_execution_skipped). Complete the skill frontmatter with the metadata fields used by cudaq-guide to address the Tier 1 static advisories. Signed-off-by: mdzurick <mitch_dz@hotmail.com>
Signed-off-by: mdzurick <mitch_dz@hotmail.com>
Disable early stopping so the NVSkills canonical Tier 3 override can run with one attempt instead of being rejected before agent evaluation. Preserve the normal two-attempt configuration and Mitch's evaluator image-mode fixes. Signed-off-by: Thomas Alexander <talexander@nvidia.com>
Signed-off-by: Thomas Alexander <talexander@nvidia.com>
taalexander
force-pushed
the
pass-agent-skill
branch
from
July 24, 2026 13:10
7c67ca3 to
7be72fd
Compare
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.
No description provided.