Skip to main content
Glama
pypi.txtβ€’1.53 kB
PyPI Packaging & Publishing (Mnemex v0.4.0) 1) Verify and edit metadata (pyproject.toml) - [project] - name = "mnemex" (if taken, fallback: "mnemex-ai" or "mnemex-memory") - version = "0.4.0" - description, readme = "README.md", license = MIT - authors, keywords set - classifiers added (Python 3.10, MIT, OS independent) - urls added: Homepage/Repository/Documentation/Issues - [tool.hatch.build] - exclude = ["marketing/**"] to keep local notes out of the sdist/wheel 2) Build and validate locally (optional) - python -m pip install --upgrade build twine - python -m build - twine check dist/* 3) Publish via CI (recommended) - A workflow is configured at .github/workflows/publish.yml to publish on tags matching v0.* - Add a repo secret: PYPI_API_TOKEN (create token on pypi.org, username: __token__, password: pypi-...) - Tag main to publish: - git tag v0.4.1 - git push origin v0.4.1 4) Manual publish (alternative) - twine upload dist/* - export TWINE_USERNAME="__token__" - export TWINE_PASSWORD="pypi-<your-token>" 5) Post-publish checks - pip install mnemex==0.4.1 (or the published version) in a fresh venv - Verify entry points work: - mnemex --help - mnemex-search --help - mnemex-maintenance --help Notes - Pre-1.0 SemVer: use 0.y.z until cross-platform/perf testing and API stability - If "mnemex" name is unavailable on PyPI at publish time, pick "mnemex-ai" and update the pyproject name accordingly before tagging. - Marketing directory is excluded; keep private notes there.

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/mnemexai/mnemex'

If you have feedback or need assistance with the MCP directory API, please join our Discord server