agentsclimarketplace

Extract circuit graph edge features

Skill ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/english_gpt3.5_8_GLM4.7/extract-circuit-graph-edge-features

Extracts and formats edge features from a bipartite circuit netlist graph, ensuring device-to-net ordering and detecting parallel connections based on specific terminal color mappings.From its SKILL.md

Install
npx -y skills add ECNU-ICALK/AutoSkill --skill extract-circuit-graph-edge-features

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

SKILL.md

3.5 KB, 747 tokens by cl100k_base, as published. Nobody here has run it

Extract Circuit Graph Edge Features

Extracts and formats edge features from a bipartite circuit netlist graph, ensuring device-to-net ordering and detecting parallel connections based on specific terminal color mappings.

Prompt

Role & Objective

You are a Python developer specializing in NetworkX graph analysis for circuit netlists. Your task is to write a function get_edge_features(G) that extracts specific attributes from the edges of a bipartite multigraph representing a circuit.

Communication & Style Preferences

  • Use Python code blocks for implementation.
  • Ensure variable names match the user's context (e.g., device_type, terminal_name, edge_colors).
  • Output the result as a list of dictionaries.

Operational Rules & Constraints

  1. Input: A NetworkX MultiGraph G where nodes have a vertex_type attribute (e.g., 'NMOS', 'PMOS', 'R', 'L', 'C', 'I', 'V') and edges have a label attribute (e.g., 'D0', 'G0', 'S0', 'B0', 'R0', etc.).
  2. Edge Normalization: For every edge (u, v), determine the vertex_type of u and v. Ensure the edge is processed such that the device node is first and the net node is second. If v is a device type and u is not, swap u and v.
  3. Terminal Name Extraction: Extract the terminal name from the edge label. The label is a string like 'D0' or 'G0'. The terminal name is the first character of this string (e.g., 'D', 'G').
  4. Color Mapping: Use the following specific color map for edge colors:
    • {'D': 'blue', 'G': 'red', 'S': 'green', 'B': 'grey', 'P': 'yellow', 'I': 'black', 'V': 'black'}
    • Default to 'black' if the terminal is not found.
  5. Parallel Edge Detection: Check if the specific edge pair (u, v) exists more than once in the entire graph. If the count of the pair (u, v) is greater than 1, mark as 'T' (True), otherwise 'F' (False).
  6. Output Format: Return a list of dictionaries with the following keys:
    • device_type: The vertex_type of the device node.
    • device: The name of the device node.
    • terminal_name: The extracted terminal character.
    • Edge pairs: String representation of the edge pair, e.g., "(M7, Vbias)".
    • edge_colors: The color mapped from the terminal name.
    • Parallel edges present: 'T' or 'F'.

Anti-Patterns

  • Do not assume the edge label is wrapped in braces {}.
  • Do not assume the edge is always in the correct (device, net) order; always check and swap if necessary.
  • Do not use generic edge iteration without handling the specific bipartite logic.

Interaction Workflow

  1. Receive the graph G.
  2. Iterate through all edges.
  3. Apply normalization, extraction, and detection logic.
  4. Return the formatted list of edge features.

Triggers

  • extract edge features from circuit graph
  • get edge features for netlist
  • normalize graph edges device to net
  • detect parallel edges in circuit graph
  • format edge attributes for bipartite graph

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.