GNCore
GNCore is a Python CLI and curated library for installing reusable AI agent skills into coding assistants.
Project version: 4.0.0 Package name: gncore License: MIT
Why it exists
AI coding assistants use different locations and formats for prompts, skills, commands and configuration. GNCore provides a common skill representation plus application adapters for installation, validation, backup, restore and updates.
Supported assistants
The current project documentation targets VS Code Copilot, Codex, and Claude harnesses, Google Antigravity, Claude Code, Gemini CLI, and OpenAI Codex CLI. Generic adapters for Claude Desktop, Cursor, Windsurf, and Continue.dev remain available for managed bundles. Hosted ChatGPT Skills can be exported as ZIP files for upload in the ChatGPT Skills UI.
Core commands
- gncore activate
- gncore deactivate
- gncore update
- gncore doctor
- gncore list
- gncore search
- gncore install
- gncore export
- gncore uninstall
- gncore backup
- gncore restore
- gncore validate
- gncore version
Useful first checks are gncore list apps, gncore validate --apps cursor and gncore doctor.
Skill model
A skill can contain metadata, a description, system prompt, examples, commands, configuration, version, dependencies and permissions.
The core pipeline is:
Skill Definition → Validation / Manifest → Adapter Selection → Application Installation → Verification → Backup / Restore / Update.
The main engineering idea is adapter isolation: adding another assistant should not require rewriting the central skill model. Native skill adapters install standard SKILL.md folders. For example:
gncore search model training
gncore install train preprocess clean-data evaluate improve analyze --apps antigravity --scope workspace
Install focused workflows such as /company-ceo, /train, /word-draft, /pdf-extract, or /ppt-create, then describe the task. Each skill guides the assistant; it does not independently execute code. GNCore installs instructions for the assistant; it does not autonomously rewrite a project. The included skills demand evidence, scoped work, data-leakage controls, and honest reporting of checks and model results.
The catalog contains 403 skills: 103 established specialist workflows plus 30 curated domain packs with 10 focused slash-invoked workflows each. Packs cover company leadership and business functions; software, data, machine learning, cloud, security, quality, and design; Word, PDF, PowerPoint, and spreadsheets; research, education, healthcare operations, retail, communications, and supply chain. Run gncore list domains, gncore list skills, or gncore search <terms> to discover workflows. Every skill carries an evidence-based inspect → plan → implement → test → fix → retest → report loop with bounded stopping criteria and honest test reporting.
Documentation
- docs/USAGE.md
- docs/ALL_SKILLS.md — all skills, use cases, slash names, and install examples
- ARCHITECTURE.md
- docs/DEVELOPER.md
- docs/API.md
- CONTRIBUTING.md
- docs/RELEASE.md
- SECURITY.md
Project status
GNCore is a real Python package and CLI under active iteration. Adapter behavior can differ between assistant versions and operating systems, so support should always be verified with the installed build.
License
MIT License. See LICENSE.
Author
Paladugu Ganesh Naidu
Repository: GNCore on GitHub
1 maintainer
Metadata
1 maintainer
Release files for gncore 4.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| gncore-4.0.0.tar.gz | 106.6 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| gncore-4.0.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 184.9 kB
Release files / gncore-4.0.0.tar.gz
| Download URL | gncore-4.0.0.tar.gz |
|---|---|
| Size | 106.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
77bf01e6742d36b8926f9bfbd32275a3281159ce0a75fe673bd1904c7b0916b1
|
|
BLAKE2b-256 checksum How to use checksums |
422592e6fd1496c7e401a18cfd04f160199e8dfd97082b027eb3de6116c98204
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Release files / gncore-4.0.0-py3-none-any.whl
| Download URL | gncore-4.0.0-py3-none-any.whl |
|---|---|
| Size | 78.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
06bee8f036894e6f66737cc2569b97bc4961ed26e0b05ee596f643030206a2bc
|
|
BLAKE2b-256 checksum How to use checksums |
825d3ef8912f40ff7283cade1fbda0d7eb0db57f7d8389fc9179c60e54690b73
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Security for gncore 4.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
