pr-review
Skill5.1k starsChanged 5 days agoLicence not identifiedView on GitHub
Review ExecuTorch pull requests for issues CI cannot check. Use when reviewing PRs, when asked to review code changes, or when the user mentions "review PR", "code review", or "check this PR".
We split this skill into the rules it gives an agent
19 rules. 0 are common, 13 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 176 edited versions we read keep these.
13
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
The licence could not be identified — read it at the source. We count its rules but don't quote them. Read them on GitHub.
“Common” is measured: kept by at least 6 in 10 of the 176 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Trending
More talk lately.about the whole repository
Who's talking about pytorch/executorch
- A Hacker News story“Executorch: On-device AI across mobile, embedded and edge for PyTorch”Hacker News · Dec 18, 2025 · 120 points
- ggml-org/llama.cpp“PR: Refine ggml-hexagon backend(Qualcomm Hexagon NPU backend) for latest ggml,whisper.cpp,llama.cpp”GitHub · Mar 11, 2025 · 13 reactions
- ByteDance-Seed/Bagel“can the Bagel supported by a very popular edge-inference engine llama.cpp?”GitHub · May 21, 2025 · 10 reactions
- ggml-org/llama.cpp“PR: Refine ggml-qnn backend(QNN, Qualcomm Neural Network,aka Qualcomm AI Engine Direct) for latest …”GitHub · Feb 24, 2025 · 9 reactions
- huggingface/transformers“Export to ExecuTorch”GitHub · Jul 26, 2024 · 5 reactions
The biggest 5 of 55 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 55
Talked about on GitHub, Hacker News, Bluesky and Stack Overflow (55 times).
On GitHub · 50
- Mar 11, 2025PR: Refine ggml-hexagon backend(Qualcomm Hexagon NPU backend) for latest ggml,whisper.cpp,llama.cpp — ggml-org/llama.cpp13 reactions · 54 comments
- May 21, 2025can the Bagel supported by a very popular edge-inference engine llama.cpp? — ByteDance-Seed/Bagel10 reactions
- Feb 24, 2025PR: Refine ggml-qnn backend(QNN, Qualcomm Neural Network,aka Qualcomm AI Engine Direct) for latest ggml,whisper.cpp,llama.cpp — ggml-org/llama.cpp9 reactions · 14 comments
- Jul 26, 2024Export to ExecuTorch — huggingface/transformers5 reactions · 17 comments
- Jul 10, 2025Add Executorch as an Official Module of Zephyr — zephyrproject-rtos/zephyr4 reactions · 12 comments
45 more on GitHub
- Dec 25, 2025use executorch for Vulkan support — resemble-ai/chatterbox2 reactions
- Jul 25, 2026Support multi-session model `LLMEngine`/`LLMSession` — abdelaziz-mahdy/executorch_flutter1 reaction · 2 comments
- Jun 4, 2026re-land :Error on raw SymInt args in fx graph call funciton nodes, and add Graph helpers to make it easier to construct correct node and materialize symints to nodes. (#183664) (#186272) — pytorch/pytorch1 reaction · 3 comments
- Jun 3, 2026chore(deps): update dependency torch to v2.7.1 [security] - autoclosed — GoogleCloudPlatform/python-docs-samples1 reaction
- Jun 2, 2026EasyCLA check should ignore Claude attributions on commits — linuxfoundation/easycla1 reaction · 1 comment
- Sep 18, 2025Add Granite Speech 3.3 — huggingface/optimum-executorch1 reaction · 1 comment
- May 14, 2025Feature request: support different input/output formats in the same recipe — meta-pytorch/torchtune1 reaction · 1 comment
- Mar 3, 2025Bias quantization for prequantized checkpoints — pytorch/ao1 reaction
- Feb 13, 2025AttributeError: type object 'torch._C._distributed_c10d.BackendType' has no attribute 'XCCL'. — pytorch/pytorch1 reaction · 13 comments
- Feb 3, 2025report regression in executorch-0.5.0 — huggingface/optimum-executorch1 reaction · 1 comment
- Jan 24, 2025Explicitly turning off pybindings for ExecuTorch install unless requested — pytorch/torchchat1 reaction · 3 comments
- Nov 13, 2024Slow down queries to docker.io — pytorch/test-infra1 reaction · 2 comments
- Oct 16, 2024Fix mobile job rate limit failures — pytorch/test-infra1 reaction · 1 comment
- Sep 27, 2024Return the list of artifacts from mobile job — pytorch/test-infra1 reaction · 1 comment
- Sep 24, 2024Set the AWS credential duration in mobile job (#5685) — pytorch/test-infra1 reaction · 1 comment
- Sep 14, 2024Add option to upload Nova artifacts to S3 directly — pytorch/test-infra1 reaction · 1 comment
- Aug 29, 2024Bring to torchtune memory wins for generate.py — meta-pytorch/torchtune1 reaction
- Apr 12, 2024Nova Linux build job runs as root inside its Docker container — pytorch/test-infra1 reaction · 7 comments
- Oct 30, 2023[dr ci] executorch docs preview link says "Python docs" when it's really all docs — pytorch/test-infra1 reaction · 1 comment
- Oct 23, 2026chore(deps): update executorch requirement from >=0.3 to >=1.5.1 in /python_executorch — FasterEdge/NPURunntime0 reactions · 1 comment
- Oct 23, 2026New test infrastructure — MDK-Packs/tensorflow-pack0 reactions
- Oct 18, 2026Review candidate: github.com/pytorch/executorch — reacher-z/Awesome-CUA0 reactions
- Oct 7, 2026Enable in-place cache updates for Whisper XNNPACK exports — huggingface/transformers0 reactions · 1 comment
- Oct 7, 2026Report the operator when ToOutVarPass meets a target it cannot handle — yenhao-huang/executorch0 reactions
- Oct 7, 2026Honour weight_n_bits so CoreML 4-bit quantization reaches PT2E — yenhao-huang/executorch0 reactions
- Oct 7, 2026Look for the torchao ops library where it is actually installed — yenhao-huang/executorch0 reactions
- Oct 7, 2026Log a summary of what a quantizer annotated — yenhao-huang/executorch0 reactions
- Oct 7, 2026Pin RoPE numerics against the textbook formula, and say what the dtype does — yenhao-huang/executorch0 reactions
- Oct 7, 2026Floor QNN per-channel weight scales at the weight width, not 16 bits — yenhao-huang/executorch0 reactions
- Oct 7, 2026Match the scalar_tensor parameter buffer to the shader declaration — yenhao-huang/executorch0 reactions
- Oct 7, 2026Hand et_vk.linear_q8ta_q8csw the float activation — yenhao-huang/executorch0 reactions
- Oct 7, 2026Report layers 8da4w skips, and let 8da8w through config validation — yenhao-huang/executorch0 reactions
- Oct 7, 2026Keep trainable parameters out of constant folding when preparing for QAT — yenhao-huang/executorch0 reactions
- Oct 7, 2026Document the pip-only AOT path for the Arm Ethos-U backend — yenhao-huang/executorch0 reactions
- Oct 7, 2026Support 5D (3D) convolution in the portable kernel — yenhao-huang/executorch0 reactions
- Oct 7, 2026Implement the portable kernel for sort.values — yenhao-huang/executorch0 reactions
- Oct 7, 2026Implement the portable kernel for scatter_reduce.two_out — yenhao-huang/executorch0 reactions
- Oct 7, 2026Implement the portable kernel for max_pool3d_with_indices.out — yenhao-huang/executorch0 reactions
- Oct 5, 2026[xnnpack] move kleidi support to opt-out (#199794) — pytorch/pytorch0 reactions · 7 comments
- Sep 15, 2026[V7 review] OpenArch – PyTorch implementations of modern LLM architectures — wsysgz/VS_AI0 reactions
- Sep 14, 2026[ATen] Handle vectorized_loop tail with masked vop, not scalar op (#197018) — pytorch/pytorch0 reactions · 11 comments
- Aug 22, 2026Executorch — seanshpark/study0 reactions · 3 comments
- Sep 4, 2025Add AArch64 unit test workflow — meta-pytorch/tokenizers0 reactions · 7 comments
- May 5, 2025Weekly pr metrics report - 2025-04-01..2025-04-07 — wdvr/pytorch0 reactions
- Apr 21, 2025Weekly pr metrics report - 2025-04-01..2025-04-07 — wdvr/pytorch0 reactions
On Hacker News · 2
- Dec 18, 2025Executorch: On-device AI across mobile, embedded and edge for PyTorch120 points · 21 comments
- May 16, 2024Executorch on-device AI across mobile, embedded and edge for PyTorch1 point · 1 comment
On Bluesky · 2
- Apr 24, 2024📦 pytorch / executorch ⭐ 712 (+16) 🗒 C++ On-device AI across mobile, embedded and edge for PyTorch — @github-trending.bsky.social2 likes and reposts
- May 2, 2024📦 pytorch / executorch ⭐ 945 (+64) 🗒 C++ On-device AI across mobile, embedded and edge for PyTorch — @github-trending.bsky.social1 likes and repost
On Stack Overflow · 1
- Dec 23, 2024cannot import name 'default_decompositions' from 'torch.export' (Pytorch & Executorch issues)0 votes · 1 answer
Other versions195 projects use a version of this.
Word for word · 0
None found.
Edited · 4
- pytorch/pytorch★ 105k49% same
- pytorch/tutorials★ 9.4k29% same
- facebookincubator/velox★ 4.2k21% same
- intel/torch-xpu-ops★ 12132% same
Wrote their own version · 191
- Significant-Gravitas/AutoGPT★ 188ksame name
- rtk-ai/rtk★ 85ksame name
- saadeghi/daisyui★ 43ksame name
- rustfs/rustfs★ 36ksame name
- dotnet/maui★ 23ksame name
- NVIDIA/Megatron-LM★ 18ksame name
- MiniMax-AI/skills★ 14ksame name
- btriapitsyn/openchamber★ 12ksame name
- openchamber/openchamber★ 12ksame name
- LMCache/LMCache★ 12ksame name
181 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
Used alongsideProjects with it also use pr-fix, clack-cli-patterns and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
volovelySkillpr-fix
This skill should be used when the user asks to "fix PR comments", "address PR feedback", "resolve PR comments", provides a GitHub PR URL with intent to fix, or says "pr-fix". It reads all review comments on a PR, decides which need code fixes vs. replies, makes the fixes, and posts replies — all with explicit user approval before any GitHub action.
Also in 5 projects that use this
jlu-lujingSkillclack-cli-patterns
Use when creating or modifying TaskHunter CLI commands, prompts, terminal output, non-TTY behavior, `--quiet`, or `--json` behavior.
Also in 3 projects that use this
- jlu-lujingSkill
drag-to-reorder
Use when implementing or modifying TaskHunter sortable or drag-to-reorder behavior, especially `@dnd-kit`, touch/mobile interactions, variable-width items, or wrapping layouts.
Also in 3 projects that use this
jtenniswoodSkillexpert-reset
Reimagine this repository from first principles using the current product and user experience as the target. Use when the user says "/expert-reset", asks how the project should be rebuilt if starting again, wants bold architecture or component-structure recommendations, or wants a research-only reset plan unconstrained by current implementation details.
Also in 3 projects that use this
CES-LtdSkilloss-pr
Use when creating a pull request, after committing changes, or when user invokes /oss-pr. Covers branch management, quality checks, commit, push, and PR creation.
Also in 3 projects that use this
- CES-LtdSkill
pr-automation
PR Automation Orchestrator: poll open PRs, check CI, run review, fix, and merge eligible PRs. Use when: (1) Invoked by daemon via scripts/pr-automation.sh, (2) User says "/pr-automation".
Also in 3 projects that use this
- +5 more from jlu-lujing
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.