read_project_artifacts

MCP tool from Marrow by desikai-lab

Reads one or more markdown artifacts

How to use it

read_project_artifacts is exposed by the Marrow MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the read_project_artifacts tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.

Install Marrow

$pip install -e .
FULL MARROW LISTING

Other tools in Marrow (22)

add_tasks

Adds a list of tasks to the project backlog

complete_tasks

Atomically closes tasks and auto-unblocks dependents

delete_project_artifact

Safely deletes an artifact

get_file_skeleton

Retrieves a token-optimized structural outline of a file

get_guideline

Assembles and returns the full context bundle (guidelines + ADRs) for any named agent role — use for mid-session role switches without disturbing pipeline state

get_project_artifact_outline

Extracts table of contents from a markdown file

get_project_map

Returns a live directory tree of all indexed files

get_session_context

Reads session state and returns phase-appropriate guidelines for the active agent role

get_task_details

Returns full task details by ID

init_project

Creates a new project workspace from the built-in template — use on Glama or any deployment without shell access

list_artifact_history

Lists version history for an artifact

list_project_artifacts

Lists files in artifact storage

list_projects

Returns a list of all available projects

move_project_artifact

Moves or renames an artifact

restore_project_artifact

Restores a previous artifact version

run_project_build

Executes a YAML build manifest to assemble context payloads

save_project_artifacts

Creates or updates artifacts (patch, replace, append)

search_code_skeletons

Semantic search over indexed source code skeletons

search_project_artifacts

Global semantic search across all artifacts

search_tasks

Semantic search over tasks

update_task

Updates task fields (status, priority, etc.)

view_file_source

Reads a precise line range from the live source repository