Albu Mcp
by dKosarevskycommunityv1.19.0
4
Model Context Protocol server for AlbumentationsX that exposes transform discovery, pipeline validation, deterministic local previews, feedback collection, and export of accepted augmentation pipelines and reports. It is a thin typed layer over AlbumentationsX primitives and is designed for bounded local filesystem preview work.
From the registry: AlbumentationsX MCP for batch previews, compare preview runs, segmentation masks, and exports.
Install & configure
Prerequisites
- requiredInstall uvInstall the uv Python package manager/runner to launch the published server with uvx.https://docs.astral.sh/uv/
- optionalOptional: prepare local image/artifact directories for preview featuresIf you want preview functionality, choose absolute filesystem paths for image access and generated artifacts. The server recommends scoping access with --allowed-root and --artifact-root.
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `albu-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install uv** — Install the uv Python package manager/runner to launch the published server with uvx. (https://docs.astral.sh/uv/)
Optional prerequisites:
- Optional: prepare local image/artifact directories for preview features — If you want preview functionality, choose absolute filesystem paths for image access and generated artifacts. The server recommends scoping access with --allowed-root and --artifact-root.
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["--from","albumentationsx-mcp","albumentationsx-mcp"]
Note: For preview work, add CLI args such as --allowed-root <absolute-path-to-images> and --artifact-root <absolute-path-to-artifacts>. Example canonical stdio args: ["--from","albumentationsx-mcp","albumentationsx-mcp","--allowed-root","<absolute-path-to-images>","--artifact-root","<absolute-path-to-artifacts>"] . No environment variables are documented in the provided README.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: For preview work, add CLI args such as --allowed-root <absolute-path-to-images> and --artifact-root <absolute-path-to-artifacts>. Example canonical stdio args: ["--from","albumentationsx-mcp","albumentationsx-mcp","--allowed-root","<absolute-path-to-images>","--artifact-root","<absolute-path-to-artifacts>"] . No environment variables are documented in the provided README.
Tools (7)
run_host_smoke_check
Runs a host smoke check to verify the MCP setup is ready for preview work.
validate_preview_request
Validates preview requests for issues like missing files, outside-root paths, masks, and annotation counts before rendering.
render_preview_batch
Renders deterministic single-image or batch previews from local images and generates contact sheets.
adjust_pipeline
Adjusts or tunes an augmentation pipeline based on preview or feedback results.
compare_preview_runs
Compares preview runs and returns a quality summary with suggested feedback tags.
export_pipeline
Exports an accepted augmentation pipeline and review artifacts.
diagnose_environment
Diagnoses preview/setup problems and returns remediation guidance.
Packages
albumentationsx-mcp
pypiv1.19.0transport: stdio
Details
- Language
- Python
- Released
- Jul 14, 2026
- Last commit
- Jun 17, 2026
- Registry name
- io.github.dKosarevsky/albu-mcp