Open-World Self-Evolution for LLM Agents — agents that build both their skills and their own verification signals from scratch, with no target-task supervision. (Code coming soon.)
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Complete pipeline for binary STL parsing, connected component extraction, volume computation, material density lookup, and mass calculation with JSON output.
Orchestrates the full pipeline from STL parsing to JSON mass report output. Coordinates parsing, component analysis, volume computation, unit conversion, and mass calculation.
Performs connected component analysis on triangle meshes using BFS with vertex adjacency and computes mesh volume via the signed tetrahedron method. Handles unit conversion and mass calculation.
Parses binary STL files using Python's struct module, extracting triangle vertices, normals, and the 2-byte attribute byte count (used as Material ID). Provides filtering by Material ID and density table parsing.
Complete Adaptive Cruise Control simulation system with PID controllers, single-loop ACC architecture (cruise/follow/emergency modes), TTC safety, Euler integration simulation, and full I/O pipeline.
Parses BibTeX files using bibtexparser 1.4.2 and cleans LaTeX/BibTeX artifacts from fields like title, author, year, DOI, and venue. Produces a normalized list of citation dictionaries ready for verification.
Compares parsed BibTeX citation metadata against API results using fuzzy string matching and multi-factor scoring to classify citations as real or fake, then outputs the final JSON report.
Inspects PDF AcroForm fields - enumerates field names, types, values, and checkbox on-states. Essential for discovering exact internal field keys before filling.
Fills PDF AcroForm fields (text and checkboxes) using pypdf, sets NeedAppearances flag, and writes the output PDF. Handles checkbox state detection and proper value updates.
Parses CIF files using pymatgen, performs space group symmetry analysis via SpacegroupAnalyzer, and extracts Wyckoff position letters, multiplicities, and representative atom fractional coordinates from the symmetrized structure.
Generates the complete visualization.js file implementing a D3.js v6 force-simulation bubble chart with sector clustering, an interactive data table, tooltips, legends, and bidirectional click-based highlighting between chart and table.
Python utility that creates the output directory structure, copies input data, downloads d3.v6.min.js locally, and writes the index.html and style.css files for the stock visualization dashboard.
Parses dialogue script text into intermediate section structures using regex, then builds a validated JSON graph schema with typed nodes and edges. Handles section headers, speaker lines, numbered choices with optional tags, and edge targets.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: