Conversation
- Add nvidia-rag-blueprint skill for NVIDIA RAG systems - Add nvidia-cuopt-numerical-optimization-api-python skill for GPU-accelerated optimization - Add nvidia-cudaq-guide skill for CUDA-Q quantum computing onboarding
Riccardo-Vecchi
left a comment
There was a problem hiding this comment.
The AST-search change needs a clean branch before technical review. This PR's merge base is 48be2e0e... (12,388 commits behind) and the comparison currently contains 59 files / +4,867 lines: the two intended AST files plus an unrelated NVIDIA optional-skill tree. The PR body is also still the untouched template, so there is no reproduction, validation, attribution/licensing note for the Cortex port, or usable test report.
Please recreate this from current main, apply only tools/ast_search.py and its focused tests, and fill in the PR description. For ported code in particular, include the exact source/revision and license compatibility so provenance can be reviewed.
|
Hi @Riccardo-Vecchi / Maintainers, I have rebased the PR onto the latest
Summary & MotivationWhen inspecting unfamiliar repositories, reading full 2,000+ line Python files wastes significant context window tokens.
This reduces LLM token usage by up to 88% during codebase exploration turns. Key Technical Design
Zero core Hermes files or workflows were modified. Ready for technical review! |
What does this PR do?
Related Issue
Fixes #
Type of Change
Changes Made
How to Test
Checklist
Code
fix(scope):,feat(scope):, etc.)pytest tests/ -qand all tests passDocumentation & Housekeeping
docs/, docstrings) — or N/Acli-config.yaml.exampleif I added/changed config keys — or N/ACONTRIBUTING.mdorAGENTS.mdif I changed architecture or workflows — or N/AFor New Skills
hermes --toolsets skills -q "Use the X skill to do Y"Screenshots / Logs