A set of Python tools to collect and summarize GitHub user activity.
This project uses uv for dependency management. It is recommended to replace python with uv run as each script is self-contained.
uv run gh_daylog.py octocat 2025-04-22
uv run gh_day_summary.py octocat-2025-04-22.ndjsonCollects all GitHub events for a specified user on a given calendar date and saves them as newline-delimited JSON.
# Set your GitHub token (optional but recommended for better rate limits)
export GH_TOKEN="ghp_xxx"
# Collect events for a user on a specific date
python gh_daylog.py octocat 2025-04-22Generates a human-readable Markdown summary from the collected events, including:
- Per-repository activity digest
- Detailed commit history with diff links
- PR and Issue interaction summary
# Generate summary from collected events
python gh_day_summary.py octocat-2025-04-22.ndjsonThe summary includes:
- Per-repo digest - Quick overview of activity counts by type for each repository
- Commit roll-up - Detailed commit history organized by repository and branch, including:
- Commit messages
- SHA hashes
- Compare diff links
- PR / Issue interaction summary - Chronological list of:
- Pull request reviews and comments
- Issue creation and comments
- Branch creation
- Other GitHub interactions
Real output for mgoin:
uv run gh_day_summary.py mgoin-2025-04-23.ndjson
# Per-repo digest
* **mgoin/hf_model_stats** – 3 × push commits
* **mgoin/llmgoin** – 2 × push commits
* **neuralmagic/vllm** – 1 × branches created, 127 × push commits
* **state-spaces/mamba** – 1 × issue opened
* **vllm-project/vllm** – 7 × reviews, 26 × review comments, 5 × issue comments, 2 × pr opened, 3 × push commits, 2 × pr closed, 1 × branches created
# Commit roll-up
## mgoin/hf_model_stats
### main
- acbbba9 Update RedHatAI stats 2025-04-23
↪︎ compare diff: https://github.com/mgoin/hf_model_stats/compare/419927ca59e005fa1b479d3e502e950e086e6b43...acbbba9cc3733e38eefd3eba1270d4e379ad76ca
- 419927c Update ibm-granite stats 2025-04-23
↪︎ compare diff: https://github.com/mgoin/hf_model_stats/compare/0f148f618a38c0ca10d27fc097962ed773db18cd...419927ca59e005fa1b479d3e502e950e086e6b43
- 0f148f6 Update nm-testing stats 2025-04-23
↪︎ compare diff: https://github.com/mgoin/hf_model_stats/compare/7a3bae6a534311ff196587b0ae25419420814115...0f148f618a38c0ca10d27fc097962ed773db18cd
## mgoin/llmgoin
### main
- 8dcd4ee Create gradio_chat.py
↪︎ compare diff: https://github.com/mgoin/llmgoin/compare/ba64213dc6d0fdf1eb07d35bcad00826c7496106...8dcd4ee066fa496defa862228944812b26a08a53
- ba64213 Create mma_test.cu
↪︎ compare diff: https://github.com/mgoin/llmgoin/compare/ef28997a5d03c665b257036f63f348266e2622bc...ba64213dc6d0fdf1eb07d35bcad00826c7496106
## neuralmagic/vllm
### test-initialization-v1
- 93e561e Improve error for structured output backend selection (#16717)
- 8a7368e [Misc] Remove redundant comment (#16703)
- 3cd91dc Help user create custom model for Transformers backend remote code models (#16719)
- 3092375 [V1][Performance] Implement custom serializaton for MultiModalKwargs [Rebased] (#16432)
- 2cbd4d2 [V1][Spec Dec Bug Fix] Respect Spec Dec Method Specification (#16636)
- 3c776dc Adding vllm buildkite job for IBM Power (#16679)
- 2b05b8c [V1][Frontend] Improve Shutdown And Logs (#11737)
- 95aca28 [rocm][V0] fix selection logic for custom PA in V0 (#16426)
- cb072ce [Bugfix] Update Florence-2 tokenizer to make grounding tasks work (#16734)
- 607029e [Bugfix] Revert max_prompt_len validation for decoder-only models. (#16741)
- 9dbf7a2 [V1] Remove log noise when idle (#16735)
- 8cac35b [Ray] Improve documentation on batch inference (#16609)
- a648152 [misc] ignore marlin_moe_wna16 local gen codes (#16760)
- 61a44a0 [Doc] Add more tips to avoid OOM (#16765)
- d8e557b [doc] add open-webui example (#16747)
- 5b1aca2 [Bugfix] Fix GLM4 model (#16618)
- 207da28 [Doc] Fix a 404 link in installation/cpu.md (#16773)
- 99ed526 [Misc] refactor examples series - lmcache (#16758)
- d27ea94 Improve configs - `TokenizerPoolConfig` + `DeviceConfig` (#16603)
- c69bf4e fix: hyperlink (#16778)
↪︎ compare diff: https://github.com/neuralmagic/vllm/compare/254d1b47a011555a53a19adc4c359cd871778bd6...ab101e2d7eb5aac371f87ec5746ab1f011a3324e
### categorize-kernel-tests
- f0b8239 Wrong location
↪︎ compare diff: https://github.com/neuralmagic/vllm/compare/6e9f325f00d476f10f1d2486364709af91f241a1...f0b82390b2de06080834aefb61c49b0dc3bad254
### enable-v1-usage-stats
- b82124d Reformat
↪︎ compare diff: https://github.com/neuralmagic/vllm/compare/24dfb08a6b130f6906ab743c47a13fe0088e24d0...b82124d21ae3e0044fe52a3e51b6d707c186569e
## vllm-project/vllm
### main
- 6317a51 Categorize `tests/kernels/` based on kernel type (#16799)
↪︎ compare diff: https://github.com/vllm-project/vllm/compare/aa72d9a4ea6b31a845bf4fbd5a97d3175a8c329a...6317a5174a4f3cbd57c44d15023042cecc576f9e
- 83d9337 [Core][V1][TPU] Enable structured decoding on TPU V1 (#16499)
↪︎ compare diff: https://github.com/vllm-project/vllm/compare/5175b884f70d82d57ad7ce5229f579e45d0c502a...83d933718c82f71e4971b6febe781743a2a52919
### tpu-v1-sampler-recompilation-test
- d7650c5 Disable enforce eager for struct_output test
↪︎ compare diff: https://github.com/vllm-project/vllm/compare/d4fb4db38e47b6beb714bcadb17cf317afd5a8bc...d7650c527d60f783a3882df436e7fcd62e03dc3f
# PR / Issue interaction summary
## neuralmagic/vllm
- Created branch `fix-missing-kernel-test`
## state-spaces/mamba
- Issue #720 OPENED: mamba-ssm fails to build on torch==2.7.0
## vllm-project/vllm
- Reviewed PR #17070 → approved
- Commented on PR #17073: Could we do this distinction based on GPU memory in GB?…
- Reviewed PR #17073 → commented
- Commented on Issue #17070: Maybe this won't work in the docker in this case because torch has not been inst…
- Commented on Issue #17070: it works if i disable build isolation i.e. `uv pip install mamba-ssm==2.2.4 --no…
- Commented on PR #17070: ```suggestion…
- Commented on PR #17070: ```suggestion…
- Reviewed PR #17070 → commented
- Commented on Issue #16859: Now that torch 2.7 has released (https://pypi.org/project/torch/2.7.0/) can this…
- PR #17060 OPENED: Add missing rocm_skinny_gemms kernel test to CI
- PR #16799 CLOSED: Categorize `tests/kernels/` based on kernel type
- Commented on PR #16850: These comments would be useful to put in the function description, if other user…
- Reviewed PR #16850 → commented
- Commented on Issue #16736: I think it is non-trivial to set the head size since this is a calculated config…
- Commented on PR #16850: Why not pass in the output here if you added the ability to do inplace?…
- Commented on PR #16850: Why can you get rid of the weight scale expand? fp8.py usually makes per-tensor …
- Commented on PR #16850: FYI @LucasWilkinson …
- Commented on PR #16850: Should we do this warning even if atomic add wasn't enabled?…
- Commented on PR #16850: So we need much smaller workspaces in all cases now? Just making sure that all t…
- Commented on PR #16850: Nice optimization for this function. This gave me an idea for an easy comment fo…
- Commented on PR #16850: Please add comments for the operations here, as this answered a question I had w…
- Reviewed PR #16850 → commented
- Commented on PR #16850: Ditto about comments on these cases just to be explicit…
- Commented on PR #16850: Did we actually support HQQ for MoE before?…
- Commented on PR #16850: cruft?…
- Commented on PR #16850: It looks like FZP_GET_IF isn't called anywhere?…
- Commented on PR #16850: Why does this file now have an entrypoint for moe marlin?…
- Commented on PR #16850: Why did you move to using auto in many places?…
- Commented on PR #16850: Can you add comments to the beginning of each of these sections? I don't underst…
- Commented on PR #16850: Why the +1?…
- Commented on PR #16850: It looks like "vllm::kU8" is included here actually so we should update the comm…
- Commented on PR #16850: This is not for the MOE generation right? So should be `MARLIN_GEN_SCRIPT`…
- Commented on PR #16850: Curious why you need cutlass_extensions?…
- Commented on PR #16850: Please add a comment for this case…
- Commented on PR #16850: Please add a comment for this case…
- Commented on PR #16850: Uncomment?…
- Commented on PR #16850: Won't `"csrc/quantization/gptq_marlin/*.cu"` include the following kernels speci…
- Reviewed PR #16850 → commented
- Reviewed PR #16850 → commented
- Commented on Issue #16986: I could not find any clear error with the failing tests other than timeouts and …
- PR #17016 OPENED: Disable enforce_eager for tests/v1/tpu/test_sampler.py
- Created branch `tpu-v1-sampler-recompilation-test`
- PR #16499 CLOSED: [Core][V1][TPU] Enable structured decoding on TPU V1
