Skip to content

Releases: m-misiura/llama-stack-provider-trustyai-fms

0.3.0

27 Nov 15:23
dd12f87

Choose a tag to compare

0.3.0

All Changes

  • 📝 bumping provider version to 0.3.0 (5fcca12)
  • 🚧 addressing breaking changes introduced by upstream PR#3797 and PR# 3895 (fa15809)
  • 🚧 Daily scheduled unit tests against llama-stack main (1fe626f)
  • 📝 updated name of gha workflows for consistency with the other providers (19d2035)
  • 🚧 run unit tests with both lls stable and main (be4a3e2)
  • 📝 change maintainer field in .toml to TrustyAI Developers (80bc3d1)
  • Rename Bandit config (2e5c33c)
  • fix: Ignore Bandit's assert warnings in tests (7959415)
  • feat: add GH workflow to run test on KinD (e456929)
  • 🚧 added a new way of loading the provider using the RemoteProviderSpec while mainting backwards compatibility (1501deb)
  • remove providers.d, parameterize module version in GHA (e569691)
  • update llama-stack version and entrypoint (46cea17)
  • feat: update pr-container-build workflow (017232d)
  • 🐛 replace non-test assertions (f23c24d)
  • 🎨 made changes to adhere to the pre-commit (largely around ruff linting and formatting) (9efba12)
  • 📝 renaming sourcery.yaml to .sourecry.yaml (f57e423)
  • 🚧 scope pre commit hooks rather than running them on all files; update hook versions (ad2eed0)
  • 🐛 fix auth token propagation in dynamic shield creation (7bfc860)
  • remove external provider dir from run.yaml (3d4c4ef)
  • add sourcery.yaml (a636ad8)
  • Remove references to providers.d (94968df)
  • remove providers.d and add ok-to-test label check (b63b4ce)
  • 🚧 addressing sourcery comments about shields caching in run_moderation and avoiding shadowing Python's built-in function input inside the method body (1237545)
  • 🚧 added two additional unit tests (empty input and single string input) (a6de132)
  • ✨ add run_moderation to the provider (5e86f97)
  • add ref to checkout step (af9cf96)
  • fix: pr-container-build workflow (078e31c)
  • 📝 updated package version and lls compatibility (307e03d)
  • feat: address LLS PR #2637 and add means of configuring provider via module (cdd6ac2)
  • Pin action SHA (993e02d)
  • chore: Add pre-commit GH action (cf991a9)
  • use sha instead of version in 3rd party images (03d6148)
  • typo (5548ac7)
  • feat: build provider distro image and add run.yaml (9c9aa7d)
  • Update python version (f5c7567)
  • Update python version (ff2ae08)
  • rename trivy job (933111c)
  • add bandit scan (02b9a4d)
  • remove empty conftest file (4d537ff)
  • fix: formatting (72345af)
  • test: add unit tests for config, detectors, and provider (473c9bf)
  • fix: docs point to upstream (4a97682)
  • 🚧 added tls support and changed from logger.info to logger.debug (f60b367)
  • 📝 changing python version to 3.12 or above (fa6ff01)
  • 🚧 dynamic shield registration via API (e651ded)
  • feat: Add GitHub Actions workflow for Trivy security scanning (87744f1)
  • ✨ added the gh workflow to automate publishing to pypi with improved performance of the workflow based on the sourcery feedback (852087b)
  • ⬇️ change to python >=3.10 in the .toml (504e0cf)
  • chore: Create dependabot.yml (47a07a6)
  • 🐛 fixing version number in pyproject.toml (ad84d03)
  • 🚧 updated pyproject.toml with license and build-system; added apache 2.0 license (dff1432)
  • 📝 updating README to indicate there are two parts: a. openshift setup; b. llama stack setup (eeddd1d)
  • 📝 updated the readme to provide more detailed instruction (7c7a112)
  • 🚧 added relevant openshift manifests for the demo (4023f2b)
  • 🚧 changed module imports and dealt with ShieldsWithACL registration bug (9991fd8)
  • 🚧 added the runtime configurations folder (3e63807)
  • 🚧 added the notebook folder (a07dc4c)
  • 📝 updating toml file to explicitly specify the package folder (8a32a87)
  • 📝 updating .gitignore to exclude python cache files and built package information (ad6d04b)
  • Delete .python-version (795c5a4)
  • Delete .ipynb_checkpoints directory (859a422)
  • 📝 change venv to .venv (2e02382)
  • 🚧 added separate shields section (04267b1)
  • 🚧 added entrypoints (bc52f8c)
  • 📝 updated port from 5001 to 8321 (26a6c19)
  • 📝 updating toml (eed8191)
  • 📝 added jupyter as a dependency as well as updated to port 8321 from 5001 (54f9209)
  • 📝 updating README (9c16e4b)
  • 🚧 added instructions on the setup and the notebook (4e27681)
  • renaming server configuration (2f11ce6)
  • 🐛 typo in a .toml file (f861555)
  • 🚧 added a configuration file together with some additional dependencies (7d25190)
  • 🎉 initial commit of the external safety provider (036f7ee)

Installation

pip install llama-stack-provider-trustyai-fms==0.3.0