Kernel Verification
ZJLi2013/awesome-kernel-skills
5-stage kernel correctness verification protocol for Triton and CUDA kernels.
Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific…
Install Autonomous GPU Kernel Timeline
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timeline --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .claude/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .claude/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timelineType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timeline --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .agents/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .agents/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timeline --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .cursor/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .cursor/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/alibaba/atrex-kernel-agent.git --path skills/autonomous-gpu-kernel-timeline--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timeline --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .gemini/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .gemini/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timelineInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .github/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .github/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install alibaba/atrex-kernel-agent autonomous-gpu-kernel-timeline --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/alibaba/atrex-kernel-agent.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/autonomous-gpu-kernel-timeline .opencode/skills/autonomous-gpu-kernel-timeline && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "autonomous-gpu-kernel-timeline" agent skill from https://github.com/alibaba/atrex-kernel-agent/tree/main/skills/autonomous-gpu-kernel-timeline into .opencode/skills/autonomous-gpu-kernel-timeline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "autonomous-gpu-kernel-timeline", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
autonomous-gpu-kernel-timeline- GitHub stars
- 154
- Token cost
- ~1.3k tokens
- SKILL.md length
- 599 words
- Files
- 8 (incl. scripts, references)
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific…
- Works in 6 steps: State one falsifiable timing question… → Save the current clean kernel.py… → Run representative correctness through… → …
- AI & LLM Engineering work in your project
- SKILL.md covers Ownership, Route, Autonomous loop and Hard boundaries
- Runs Python scripts from its folder
What it does
Autonomous GPU Kernel Timeline is an agent skill from alibaba/atrex-kernel-agent. Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific kernel-internal timing question.
Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 12 other files, including scripts and reference files (for example `backends/cuda_backend/adapter.py`, `backends/cutedsl_backend/adapter.py` and `references/cuda-backend.md`).
It sits in AI & LLM Engineering. It works with CUDA. The repository describes itself as: An end-to-end agent project for GPU kernel implementation, analysis, profiling, and iterative optimization. It helps an agent turn PyTorch logic or an existing kernel into a… The licence is Apache-2.0.
When your agent uses it
- AI & LLM Engineering work in your project
Example prompts
- “/autonomous-gpu-kernel-timeline”
Requirements
- Python 3
Workflow steps
6 steps, taken from the first numbered list in SKILL.md.
- State one falsifiable timing question and begin with the fewest useful sites.
- Save the current clean kernel.py content, then create an instrumented working snapshot in the
- Run representative correctness through the immutable evaluator and capture through the campaign
- Read the summary and Perfetto trace. Remove, move, or refine probes and repeat when the evidence
- Before replacing the snapshot, preserve its source or reversible patch and its content hash.
- Restore a probe-free kernel, implement the optimization, and use the normal evaluator for final
What it can do on your machine
Read from SKILL.md and the folder at commit 3d27c1e. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/ (Python), which the agent can run.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Autonomous GPU Kernel Timeline loads about 1.3k tokens when it runs, and up to ~3.7k if it reads all its reference files. Until then it costs about 65 tokens; SKILL.md has 599 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from alibaba/atrex-kernel-agent at commit 3d27c1e, republished under its Apache-2.0 licence (© alibaba). 599 words, ~1,250 tokens.
Download SKILL.mdSave it as
.claude/skills/autonomous-gpu-kernel-timeline/SKILL.md (or your agent's skills folder). This skill also uses 7 other files; get the full folder from GitHub.- name
- autonomous-gpu-kernel-timeline
- description
- Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific kernel-internal timing question.
Autonomous GPU Kernel Timeline
Use this inside an AKA optimization episode only after a correct runnable kernel and representative workload exist. Do not trigger it when aggregate kernel timing or ordinary profiler evidence already answers the question.
Ownership
- AKA chooses the hypothesis, sites, writer roles, density, boundary semantics, and number of retries.
- The backend records and decodes those choices without silent identity, capacity, or pairing errors.
- The final validator rejects known factual failures; it does not decide whether AKA chose the best algorithm boundary or interpretation.
Route
- Standalone CUDA and inline PTX: read references/cuda-backend.md
and use
backends/cuda_backend/atrex_timeline.cuh. - CuTe DSL: read references/iket-quickstart.md and use IKeT. Do not fall back to the CUDA backend.
Autonomous loop
- State one falsifiable timing question and begin with the fewest useful sites.
- Save the current clean
kernel.pycontent, then create an instrumented working snapshot in the same episode worktree. This snapshot is not a handoff candidate and must not enter promotion or stall accounting. - Run representative correctness through the immutable evaluator and capture through the campaign
sandbox. For final evidence, pass the sandbox-owned
.atrex_long_horizon/evaluations.jsonlas--correctness-evidence; a model-supplied--correctness passedis only an exploration note and cannot producedecision_grade. Usescripts/timeline.pyto validate and export the returned evidence. If the remote command reads backend files, pass--input skills/autonomous-gpu-kernel-timeline/backends/<backend>totools/sandbox.py; sync only the attempt-specific directory. - Read the summary and Perfetto trace. Remove, move, or refine probes and repeat when the evidence
is insufficient, semantically misplaced, or too intrusive. Name every reported interval as
start_site -> end_siteand recompute its numbers from canonical events rather than trusting a prose label. A local timeline delta is mechanism evidence, not a substitute for probe-free end-to-end ABBA. Reviewer feedback is optional. - Before replacing the snapshot, preserve its source or reversible patch and its content hash.
- Restore a probe-free kernel, implement the optimization, and use the normal evaluator for final correctness and performance. Re-instrument the new clean state only when confirmation is useful.
For a final perturbation claim, use scripts/timeline.py measure inside one GPU allocation. Give it
baseline/instrumented commands as JSON argv arrays, the exact sources, and materialized binaries when
the compiler exposes them. JIT-only CuTe DSL measurements need not invent a binary artifact.
Each command must perform the requested warmup and iterations, synchronize the device, check the full
representative output, and emit exactly one line of this form:
Show full SKILL.md (206 more words)Show less
__ATREX_TIMELINE_SAMPLE__={"latency_ms":1.0,"correctness":"passed","synchronized":true,"workload_identity":"...","device_identity":{"uuid":"..."},"warmup":10,"iterations":100}The helper runs each sample in a fresh process, defaults to ABBA followed by BAAB, rejects workload or
device drift, and writes the raw schedule and samples. Pass that artifact to capture/export with
--measurement; validate rechecks its hashes and recomputes the medians and overhead.
The sample's correctness field rejects a bad timing run but does not replace the immutable evaluator
record required for final evidence.
Hard boundaries
- Never modify
profile_driver.py, evaluators, ground truth, or other protected paths. - Never hand off or promote an instrumented snapshot. Only a probe-free kernel may become the
episode
candidate_commit == HEAD. - Do not combine events from different launches into one apparent execution.
- Construct exactly one recorder per selected owner per launch and reuse it for that owner's events; duplicate ownership is a capture failure, not a sampling policy.
- Reject overflow, truncation, invalid owner/site identity, deterministic range mismatch, stale source/binary/workload provenance, and unrecomputable numeric claims.
- Never classify final evidence as
decision_gradewithout a matching sandbox evaluator record. - The default low-perturbation target is 10%. A higher-overhead trace may remain diagnostic when it is labelled honestly; the threshold is hard only when the task explicitly requires it.
- Do not add fixed site catalogs, fixed coverage quotas, mandatory reviewer approval, or a separate orchestration state machine.
© alibaba, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 7 other files (scripts, references) in skills/autonomous-gpu-kernel-timeline of alibaba/atrex-kernel-agent.
- SKILL.md
- backends/cuda_backend/adapter.py
- backends/cuda_backend/atrex_timeline.cuh
- backends/cuda_backend/test_backend.cu
- backends/cutedsl_backend/adapter.py
- references/cuda-backend.md
- references/iket-quickstart.md
- scripts/timeline.py
Open the folder on GitHubat commit 3d27c1e
Compare with similar skills
Autonomous GPU Kernel Timeline next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Autonomous GPU Kernel Timeline this skillalibaba/atrex-kernel-agent | 154 | — | ~1.3k | Automated safety check: Pass | Apache-2.0 | |
| Kernel VerificationZJLi2013/awesome-kernel-skills | 102 | — | ~702 | Automated safety check: Pass | None | |
| Esmfold2JimLiu/science-skills | 227 | 4 repos | ~2.5k | Automated safety check: Pass | Apache-2.0 | |
| Hugging Face Local Modelshuggingface/skills | 11k | 3 repos | ~945 | Automated safety check: Pass | Apache-2.0 | |
| Setup Workshop Nemoclawbrevdev/workshop-build-an-agent | 143 | — | ~5.2k | Automated safety check: Pass | Apache-2.0 | |
| Make Op VerifyCVCUDA/CV-CUDA | 2.7k | — | ~433 | Automated safety check: Pass | Custom licence |
Similar skills
Esmfold2
JimLiu/science-skills
Biohub ESMFold2 / ESMFold2-Fast all-atom co-folding (Candido et al.
- Official
Hugging Face Local Models
huggingface/skills
Finds llama.cpp-compatible GGUF models on the Hugging Face Hub, picks a quantization for your hardware and launches them with llama-cli or llama-server.
Setup Workshop Nemoclaw
brevdev/workshop-build-an-agent
Set up the NVIDIA "Build an Agent" DevX workshop as a working JupyterLab environment from INSIDE a locked-down OpenShell/NemoClaw sandbox, and hand the user the token URL + access commands.
Make Op Verify
CVCUDA/CV-CUDA
Verify a new CV-CUDA operator against the deterministic final regression checklist (the /make-op done-gate).
GPU Optimizer
Mathews-Tom/armory
GPU optimization for consumer NVIDIA GPUs (8-24GB VRAM) covering mixed precision, gradient checkpointing, XGBoost GPU, CuPy/cuDF migration, and torch.compile.
More from alibaba/atrex-kernel-agent
Opt Trace Mining
alibaba/atrex-kernel-agent
Mine a per-kernel optimization trace — a git repository capturing successive versions of one kernel being optimized — into structured, gate-validated optimization-experience records for the GPU…
Session Trace Mining
alibaba/atrex-kernel-agent
Mine AI coding-agent session transcripts into structured, gate-validated GPU-kernel optimization records for the wiki.
Ppu Acu Joint Profile
alibaba/atrex-kernel-agent
Choose and run ACU-only, adaptive PPU in-kernel timeline, or optional bounded joint analysis for a PPU kernel.
Gen Plan
alibaba/atrex-kernel-agent
Generate a structured implementation plan from an evidence draft.
GPU Kernel Baseline
alibaba/atrex-kernel-agent
Learn the target framework from enabled knowledge tools and implement a baseline GPU kernel.
GPU Kernel Episode Loop
alibaba/atrex-kernel-agent
Run the evidence loop of one long-horizon GPU kernel optimization episode.
Related
Works with
Categories
Questions about Autonomous GPU Kernel Timeline
What does Autonomous GPU Kernel Timeline do?
Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific…. Autonomous GPU Kernel Timeline is an agent skill from alibaba/atrex-kernel-agent. Let AKA autonomously add, run, inspect, and revise intra-kernel timeline probes for standalone CUDA/inline PTX or CuTe DSL when ordinary benchmark, NSYS, or NCU evidence cannot answer a specific kernel-internal timing question.
When should I use Autonomous GPU Kernel Timeline?
Autonomous GPU Kernel Timeline fits situations like: AI & LLM Engineering work in your project.
How do I install Autonomous GPU Kernel Timeline in Claude Code?
Run `npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a claude-code`. Or copy the skill folder (skills/autonomous-gpu-kernel-timeline in alibaba/atrex-kernel-agent) into .claude/skills/autonomous-gpu-kernel-timeline in your project. Claude Code loads it when a task matches its description.
How do I install Autonomous GPU Kernel Timeline in Codex?
Run `npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a codex`. Or copy the skill folder (skills/autonomous-gpu-kernel-timeline in alibaba/atrex-kernel-agent) into .agents/skills/autonomous-gpu-kernel-timeline in your project. Codex loads it when a task matches its description.
Can I use Autonomous GPU Kernel Timeline in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add alibaba/atrex-kernel-agent --skill autonomous-gpu-kernel-timeline -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/autonomous-gpu-kernel-timeline, .gemini/skills/autonomous-gpu-kernel-timeline, .github/skills/autonomous-gpu-kernel-timeline and .opencode/skills/autonomous-gpu-kernel-timeline in your project.
What does Autonomous GPU Kernel Timeline need to run?
Going by SKILL.md and its folder, Autonomous GPU Kernel Timeline needs Python for the scripts in its folder. Our summary lists: Python 3.
Does Autonomous GPU Kernel Timeline access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Autonomous GPU Kernel Timeline safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Autonomous GPU Kernel Timeline use?
Autonomous GPU Kernel Timeline is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Autonomous GPU Kernel Timeline use?
About 1.3k tokens (SKILL.md is roughly 5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2.4k tokens, read only when the agent opens those files.
What are the alternatives to Autonomous GPU Kernel Timeline?
Skills that share tags, products or a category with Autonomous GPU Kernel Timeline: Kernel Verification (ZJLi2013/awesome-kernel-skills, 102 stars), Esmfold2 (JimLiu/science-skills, 227 stars), Hugging Face Local Models (huggingface/skills, 11k stars) and Setup Workshop Nemoclaw (brevdev/workshop-build-an-agent, 143 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Autonomous GPU Kernel Timeline?
alibaba (a GitHub organization) maintains it in alibaba/atrex-kernel-agent, which has 154 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on September 29, 2026.
Source: alibaba/atrex-kernel-agent on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.