New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Graph Layers

by VectorSpaceLab in VectorSpaceLab/AREX-Skill

Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API…

MITAuto-check passed

Install Graph Layers

skills CLI
$ npx skills add VectorSpaceLab/AREX-Skill --skill graph-layers -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install VectorSpaceLab/AREX-Skill graph-layers --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/VectorSpaceLab/AREX-Skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/repositories/repo-skills/deep-gcns-torch/sub-skills/graph-layers .claude/skills/graph-layers && rm -rf skills-src

Use ~/.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/

Facts

Skill name
graph-layers
GitHub stars
328
Token cost
~1.2k tokens
SKILL.md length
509 words
Files
6 (incl. scripts, references)
Skills in repo
157
Repo updated
First seen
Licence
MIT

At a glance

Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API…

  • Works in 7 steps: Identify layout. Use sparse node… → Choose graph construction. Use a… → Choose convolution. Sparse GraphConv… → …
  • Layer-level API and shape questions
  • SKILL.md covers Route by task, Quick decision procedure, Dependency boundary and References
  • Runs Python scripts from its folder; calls python and pip

What it does

Graph Layers is an agent skill from VectorSpaceLab/AREX-Skill. Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API and shape questions, not end-to-end dataset workflows.

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 7 other files, including scripts and reference files (for example `references/aggregation-and-blocks.md`, `references/api-reference.md` and `references/reversible.md`).

The repository describes itself as: A Skill Library for Automated Machine Learning. The licence is MIT.

When your agent uses it

  • Layer-level API and shape questions
  • Not end-to-end dataset workflows

Example prompts

  • “/graph-layers”

Requirements

  • Python 3

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Identify layout. Use sparse node features (N, C) with a PyG
  2. Choose graph construction. Use a supplied edge_index when topology is
  3. Choose convolution. Sparse GraphConv supports edge, mr, gat,
  4. Choose composition. Plain blocks transform features, residual blocks
  5. For generalized aggregation, configure GENConv and validate the
  6. For memory-efficient depth, use a channel-divisible group additive
  7. Run the safe helper from any working directory

What it can do on your machine

Read from SKILL.md and the folder at commit ac3fe1a. 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.

    Shell commands in SKILL.md call:

    • python
    • pip

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use pip, which can reach the network depending on how they are called.

    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

Graph Layers loads about 1.2k tokens when it runs, and up to ~6.4k if it reads all its reference files. Until then it costs about 66 tokens; SKILL.md has 509 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~66
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~6.4k

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 VectorSpaceLab/AREX-Skill at commit ac3fe1a, republished under its MIT licence (© VectorSpaceLab). 509 words, ~1,173 tokens.

Download SKILL.mdSave it as .claude/skills/graph-layers/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
graph-layers
description
Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API and shape questions, not end-to-end dataset workflows.
disable-model-invocation
true
metadata.disco-role
operating
license
MIT

Graph layers

Use this skill for a layer-level design or debugging task. It is a distilled operating guide, not a source-checkout import recipe. Start with the tensor layout and graph representation, choose a static or dynamic layer, then run the small bundled smoke before moving to a real workload.

Route by task

  • For ModelNet40, S3DIS, PartNet, point-cloud data loading, task flags, checkpoints, or visualization, hand off to the sibling point-cloud-workflows skill.
  • For OGB datasets, DeeperGCN benchmark configurations, graph pooling, partitioning, or RevGNN/RevGAT experiments, hand off to ogb-workflows.
  • For PPI data, F1 metrics, and PPI training/evaluation, hand off to ppi-workflows.
  • Keep the root skill responsible for installation and broad routing. This skill can identify dependency failures but does not install packages or run long training.

Quick decision procedure

  1. Identify layout. Use sparse node features (N, C) with a PyG edge_index (2, E) and optional batch (N,) for independent graphs; use dense point-cloud features (B, C, N, 1) with dense indices (2, B, N, K). Do not pass a dense tensor to sparse layers or flatten a dense batch without preserving graph membership.

  2. Choose graph construction. Use a supplied edge_index when topology is fixed. Otherwise use DynConv/DynConv2d, selecting kernel_size K, dilation d, and (for dense layers) knn='matrix' or the compiled torch_cluster path. Ensure K*d does not exceed points per graph.

  3. Choose convolution. Sparse GraphConv supports edge, mr, gat, gcn, and gin in the inspected implementation. sage and rsage are exposed but are not a supported modern-PyG route; see troubleshooting. Dense GraphConv2d and DynConv2d support edge and mr.

  4. Choose composition. Plain blocks transform features, residual blocks add a same-width scaled skip, and dense blocks concatenate newly produced channels. Static blocks preserve and return edge_index; sparse dynamic blocks return (features, batch).

  5. For generalized aggregation, configure GENConv and validate the aggregator, temperature/power parameters, edge encoding, and message normalization together. See aggregation and blocks.

  6. For memory-efficient depth, use a channel-divisible group additive coupling with deterministic per-group functions, then wrap it with the reversible wrapper only after a direct forward/inverse round trip passes. See reversible.

  7. Run the safe helper from any working directory:

    bash
    python /absolute/path/to/graph-layers/scripts/layer_smoke.py --help
python /absolute/path/to/graph-layers/scripts/layer_smoke.py --tiny

    Resolve the absolute path in the caller's skill installation; never add a source checkout to PYTHONPATH for this helper.

Show full SKILL.md (137 more words)Show less

Dependency boundary

The core layer behavior depends on a coherent PyTorch, PyTorch Geometric, torch-scatter, and torch-cluster installation. The verified inspection combination was PyTorch 2.11.0+cu128, PyG 2.8.0.post1, torch-scatter 2.1.2+pt211cu128, and torch-cluster 1.6.3+pt211cu128, with pip check passing. Treat those versions as evidence, not as a universal pin: match PyG extension wheels to the installed PyTorch and CUDA/CPU build. A missing or ABI-incoherent compiled extension is a dependency failure, not a layer-shape bug. The tiny helper probes these dependencies without importing repository modules.

Exact benchmark reproduction is intentionally outside this skill. The repository-era PyG 1.6.3 probe did not import with torch 1.13.1 because of the removed torch._six.container_abcs; use a coherent historical environment for old-number reproduction and do not infer benchmark equivalence from the modern smoke.

References

© VectorSpaceLab, MIT. 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 5 other files (scripts, references) in skills/repositories/repo-skills/deep-gcns-torch/sub-skills/graph-layers of VectorSpaceLab/AREX-Skill.

  • SKILL.md
  • references/aggregation-and-blocks.md
  • references/api-reference.md
  • references/reversible.md
  • references/troubleshooting.md
  • scripts/layer_smoke.py

Open the folder on GitHubat commit ac3fe1a

Compare with similar skills

Graph Layers 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.

Graph Layers compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Graph Layers this skillVectorSpaceLab/AREX-Skill328—~1.2kAutomated safety check: PassMIT
Graphagenticnotetaking/arscontexta3.5k1 repos~4.9kAutomated safety check: NotesMIT
Bigquery Graphgoogle/adk-python22k—~4.8kAutomated safety check: PassApache-2.0
Code Review Graph Buildertirth8205/code-review-graph32k—~295Automated safety check: PassMIT
Mini Context Graphgithub/awesome-copilot40k1 repos~2kAutomated safety check: PassMIT
Graphatopile/atopile4k—~952Automated safety check: PassMIT

Similar skills

  • Graph

    agenticnotetaking/arscontexta

    Interactive knowledge graph analysis. An agent skill from agenticnotetaking/arscontexta.

    3.5k GitHub starsUsed in 1 repo~4.9k tokens
    Knowledge ManagementAuto-check: notes

  • Bigquery Graph

    google/adk-python

    Official

    Skill for Graph Query Language (GQL) or SQL/PGQ queries against a property graph.

    22k GitHub stars~4.8k tokensUpdated today
    DatabasesAuto-check passed

  • Code Review Graph Builder

    tirth8205/code-review-graph

    Builds or incrementally updates the code-review knowledge graph for a repository and reports whether the build succeeded, partly or failed.

    32k GitHub stars~295 tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Mini Context Graph

    github/awesome-copilot

    Official

    A persistent, compounding knowledge base combining Karpathy's LLM Wiki pattern with a structured knowledge graph.

    40k GitHub starsUsed in 1 repo~2k tokens
    Knowledge ManagementAuto-check passed

  • Graph

    atopile/atopile

    How the Zig-backed instance graph works (GraphView/NodeReference/EdgeReference), the real Python API surface, and the invariants around allocation, attributes, and cleanup.

    4k GitHub stars~952 tokensUpdated 3 mo ago
    Auto-check passed

  • Social Graph Ranker

    affaan-m/ECC

    Weighted social-graph ranking for warm intro discovery, bridge scoring, and network gap analysis across X and LinkedIn.

    275k GitHub starsUsed in 1 repo~1k tokens
    Auto-check passed

More from VectorSpaceLab/AREX-Skill

All 157 skills in this repo

  • Agent Lightning

    VectorSpaceLab/AREX-Skill

    Use this repo skill for Agent Lightning package tasks: authoring trainable agents, tracing rewards and spans, running LightningStore/Trainer loops, using agl CLI services, choosing examples, and…

    328 GitHub stars~1.4k tokensUpdated 1 mo ago
    Auto-check passed

  • Agent Tools

    VectorSpaceLab/AREX-Skill

    A skill your agent uses when configuring LiteLLM for MCP tools, A2A agents, Claude Code/Cursor agent gateway traffic, MCP auth/OAuth, tool permissions, semantic filtering, or agent-specific proxy…

    328 GitHub stars~1.2k tokensUpdated 1 mo ago
    Auto-check passed

  • Agents And Awel

    VectorSpaceLab/AREX-Skill

    Build and debug DB-GPT agents, tools, skills, teams, and AWEL workflows, including deterministic local DAG runs and HTTP-trigger topology without assuming an LLM, credential, or external service.

    328 GitHub stars~2.3k tokensUpdated 1 mo ago
    Auto-check passed

  • Agents And Middleware

    VectorSpaceLab/AREX-Skill

    Work on the actively maintained LangChain v1 agent package: initchatmodel, createagent, structured output, tools, middleware, embeddings initialization, provider routing, and agent runtime…

    328 GitHub stars~1.2k tokensUpdated 1 mo ago
    Auto-check passed

  • Agents Workflows

    VectorSpaceLab/AREX-Skill

    A skill your agent uses for giskard.agents async chat workflows, tools, prompt templates, structured outputs, retries, rate limiting, embeddings, and optional LiteLLM backend.

    328 GitHub stars~500 tokensUpdated 1 mo ago
    Auto-check passed

  • Alphafold3

    VectorSpaceLab/AREX-Skill

    A skill your agent uses for AlphaFold 3 input preparation, prediction command planning, output interpretation, and Python API inspection.

    328 GitHub stars~1.2k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Graph Layers

What does Graph Layers do?

Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API…. Graph Layers is an agent skill from VectorSpaceLab/AREX-Skill. Construct and troubleshoot the repository's sparse and dense graph layers, dynamic or dilated KNN blocks, GENConv aggregation, and reversible coupling primitives; use this skill for layer-level API and shape questions, not end-to-end dataset workflows.

When should I use Graph Layers?

Graph Layers fits situations like: layer-level API and shape questions; not end-to-end dataset workflows.

How do I install Graph Layers in Claude Code?

Run `npx skills add VectorSpaceLab/AREX-Skill --skill graph-layers -a claude-code`. Or copy the skill folder (skills/repositories/repo-skills/deep-gcns-torch/sub-skills/graph-layers in VectorSpaceLab/AREX-Skill) into .claude/skills/graph-layers in your project. Claude Code loads it when a task matches its description.

How do I install Graph Layers in Codex?

Run `npx skills add VectorSpaceLab/AREX-Skill --skill graph-layers -a codex`. Or copy the skill folder (skills/repositories/repo-skills/deep-gcns-torch/sub-skills/graph-layers in VectorSpaceLab/AREX-Skill) into .agents/skills/graph-layers in your project. Codex loads it when a task matches its description.

Can I use Graph Layers 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 VectorSpaceLab/AREX-Skill --skill graph-layers -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/graph-layers, .gemini/skills/graph-layers, .github/skills/graph-layers and .opencode/skills/graph-layers in your project.

What does Graph Layers need to run?

Going by SKILL.md and its folder, Graph Layers needs Python for the scripts in its folder and the command-line tools its instructions call (python and pip). Our summary lists: Python 3.

Does Graph Layers access the network?

SKILL.md contains no URLs. Its commands use pip, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Graph Layers 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 Graph Layers use?

Graph Layers is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Graph Layers use?

About 1.2k tokens (SKILL.md is roughly 4.7k 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 5.2k tokens, read only when the agent opens those files.

What are the alternatives to Graph Layers?

Skills that share tags, products or a category with Graph Layers: Graph (agenticnotetaking/arscontexta, 3.5k stars), Bigquery Graph (google/adk-python, 22k stars), Code Review Graph Builder (tirth8205/code-review-graph, 32k stars) and Mini Context Graph (github/awesome-copilot, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Graph Layers?

VectorSpaceLab (a GitHub organization) maintains it in VectorSpaceLab/AREX-Skill, which has 328 GitHub stars. The repository holds 157 skills in this directory. The repository was last updated on September 3, 2026.

Source: VectorSpaceLab/AREX-Skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.