Plane Workflow MCP
Plane Workflow MCP lets an agent inspect, plan, and safely apply Plane work-item workflows from local git history and structured project operations.
Check setup, list/activate workspaces and projects, and inspect the active context.
Read project workflow profiles/options and validate or save local profile overrides.
Create and review read-only standardization plans, then apply them only with confirmation.
Find duplicate candidates and work items, list/filter work items, get project briefings, audit work items, and export DOCX/PDF reports.
Manage work-item relations, evidence links, attachments, and modules.
Create, start, complete, cancel, and update structured work items with planning fields, duplicate checks, idempotency, and dry-run previews.
Plan git backfills from commit history, review proposed items, then apply them to create/start/complete items with commit-citing comments and resumable idempotent writes.
Write operations are tied to the active workspace/project unless another project is explicitly confirmed, with previews and safety checks.
Reads local git history and logs, groups commits into Plane work items, and backfills planned, started, and completed items with commit citations.
Provides an optional Mattermost bot integration for using the Plane workflow from Mattermost.
Plane Workflow MCP
Your git history already knows what you did. Let your agent put it in Plane.
An open-source MCP server and CLI for Plane, cloud or self-hosted. It reads your commits and turns them into Plane work items that are planned, started, and completed on the right business days, with notes that cite the commits. You see the whole plan first; nothing is written until you approve it.

Offline demo: the real server code runs against a test fixture with synthetic data. Reproduce it.
Who it is for: teams on Plane who work through Claude Code, Codex, Cursor, or another MCP client, and whose board falls behind what they actually shipped. It works next to the official Plane MCP server rather than replacing it (how they differ).
Quick start
Requires Python 3.11+, uv, and a Plane API key.
uv tool install plane-workflow-mcp
plane-workflow setup --client claude-codeSetup asks for your Plane URL, workspace, and API key. It checks the connection,
stores the key in your OS keyring, and registers the server with your client. Other
clients: --client codex, cursor, claude-desktop, opencode, or zed; any other
stdio client can run plane-workflow mcp. Add --dry-run to see the change first.
Using Claude Code? You can install it as a plugin instead, which also adds a skill for using the tools safely.
Select a project, then restart your client:
plane-workflow project list
plane-workflow project activate PROJECT_IDThen preview a backfill from any of your repositories. Nothing is written to Plane
without --apply:
plane-workflow backfill --since 2026-09-01Update
plane-workflow status shows your version and tells you when a newer release
exists. Update with uv tool install --force plane-workflow-mcp, then restart your client. Your
setup, API key, and selected project are kept, so there is no need to run setup again.
Related MCP server: claude-plane
What it does
Backfill Plane from git.
plan_git_backfillreadsgit logand groups commits into work items by module, scope, or folder. Dates come from the commits and are moved onto your business days. Nothing is written until you approve the plan.apply_git_backfillthen creates, starts, and completes each item, with a comment listing its commits. Re-running it never duplicates items. Also available asplane-workflow backfill --since 2026-09-01.Plan → start → complete, with evidence. New work gets scope, acceptance criteria, assignee, estimate, and start/target dates that skip your non-business days. Completion requires a started item, a factual summary, and real verification. It never fabricates either.
Preview first, safe to retry. Write tools support
dry_runpreviews. Creates check for duplicates and are idempotent, so a timeout followed by a retry does not create a second item. Rate limits are retried with backoff.Know what needs attention. Read-only briefings list overdue, stale, unassigned, unestimated, and unscheduled work. Reports export to DOCX or PDF.
Never touch the wrong project. Writes are tied to the selected workspace and project, unless one call explicitly confirms another project's identifier.
First prompt
Read this repository's git history since September 1 and propose Plane work items
for it with plan_git_backfill. Show me the plan as a table. Do not write anything
to Plane until I approve.Or start read-only:
Use Plane Workflow to confirm my active workspace and project, then show overdue
and unassigned work in that project with identifiers and totals. Do not change
any Plane data.Then try a write with a preview:
Retrying checkout can create two orders. Draft a bug with acceptance criteria in
the active project using its workflow rules. Show the dry-run preview and wait for
my approval before creating it.More prompts with expected results · Run the offline demo
How it compares
The official Plane MCP server gives broad access to Plane resources and hosted OAuth. This server adds a workflow layer on top of Plane: planning rules, an enforced lifecycle, previews, idempotent writes, and completion records. You can install both. Detailed comparison.
Review happens in your client. The server enforces planning rules and lifecycle checks, but it cannot prove a human approved a call. Plane is updated when the agent calls a tool; there is no background sync.
Documentation
MIT licensed. Community project, not affiliated with Plane.
If it saves your team a round of status updates, a star helps other Plane users find it.
Available Tools
33 toolsactivate_plane_projectB
Validate and select a project in the active stored workspace for future work-item operations.
Inputs: project_id (required). Returns: object
activate_plane_workspaceA
Switch to a configured Plane workspace profile without contacting Plane or exposing its API key.
Inputs: profile (required). Returns: object
add_work_item_evidence_linksB
Preview or attach validated evidence links to one work item. Set confirm=true to create the links.
Inputs: note, confirm, evidence, project_id, +2 more. Returns: object
add_work_item_relationA
Preview or create same-project work-item relations. Set confirm=true after reviewing the relation direction.
Inputs: work_item_id (required), relation_type (required), related_work_item_ids (required), confirm, +2 more. Returns: object
apply_git_backfillA
Preview or apply a saved git backfill plan: create, start, and complete each item with a commit-citing record.
titles and modules override proposed values by item id; skip leaves items out. Committed work is completed with a comment listing its commits; uncommitted work is only started. Every step is idempotent, so re-running after an error resumes without duplicates. Plane sets completed_at to the time of this call; the real dates are kept in start_date, target_date, and the comment.
Inputs: plan_id (required), skip, titles, dry_run, +3 more. Returns: object
apply_standardization_planA
Preview or apply a saved standardization plan. Set confirm=true only after reviewing its proposed actions.
Inputs: plan_id (required), confirm. Returns: object
audit_work_itemsB
Audit Plane work items without changing them. Reports structural and quality gaps for a human-approved cleanup.
Inputs: project_id. Returns: object
cancel_standard_work_itemB
Preview or cancel unfinished work after recording a factual reason and optional follow-ups.
Inputs: reason (required), work_item_id (required), dry_run, state_id, +3 more. Returns: object
complete_standard_work_itemC
Preview or complete started work after recording a factual, human-readable completion comment.
Inputs: summary (required), verification (required), work_item_id (required), dry_run, +6 more. Returns: object
create_standardization_planB
Create and save a read-only cleanup proposal. The plan does not change Plane until apply_standardization_plan is confirmed.
Inputs: project_id, expires_in_hours. Returns: object
create_standard_work_itemB
Create a structured Plane work item with labels, duplicate review, optional module/cycle assignment, and planning fields.
Retrying a create with the same idempotency_key (or, without one, the same title) returns the item the first attempt made instead of creating a second one. Use distinct idempotency_keys with allow_duplicate=true to create intentionally separate items that share a title.
Inputs: outcome (required), acceptance_criteria (required), scope, context, +23 more. Returns: object
diagnose_plane_connectionA
Check local Plane configuration and, when given a project, read the available API capabilities without changing Plane.
Inputs: project_id. Returns: object
ensure_moduleC
Preview or find an existing module, then explicitly create or assign it when requested.
Inputs: dry_run, project_id, description, module_name, +3 more. Returns: object
export_work_items_reportB
Export filtered Plane work items as a dynamic Word or PDF report without changing Plane.
Example filters: {"state_names": ["Backlog", "In Progress"]}. Example layout: {"group_by": "state", "columns": ["identifier", "title", "state", "priority"]}.
Inputs: format (required), title, layout, filters, +2 more. Returns: object
find_duplicate_candidatesC
Find potentially duplicate Plane work items by title similarity without changing Plane.
Inputs: title, min_score, project_id, max_results. Returns: object
find_work_itemsB
Find Plane work items by their title, UUID, or project reference such as PROJECT-12 without changing Plane.
Inputs: query, project_id, max_results. Returns: object
get_active_plane_contextA
Show the active Plane workspace and project without exposing credentials or changing Plane.
Inputs: no named parameters. Returns: object
get_plane_workflow_setup_statusA
Check whether Plane Workflow is configured, without contacting Plane or revealing credentials.
Inputs: no named parameters. Returns: object
get_project_briefingB
Summarize project execution and identify overdue, stale, unassigned, and unplanned active work.
Inputs: project_id, attention_limit, stale_after_days. Returns: object
get_project_workflow_contextB
Read a project's identifiers, active profile, modules, labels, and work-item count without changing Plane.
Inputs: project_id. Returns: object
get_project_workflow_profileB
Read the merged workflow profile for a Plane project without changing Plane or local profile files.
Inputs: project_id. Returns: object
get_standardization_planB
Read a saved standardization plan without changing Plane.
Inputs: plan_id (required). Returns: object
get_workflow_optionsB
List the available state, cycle, assignee, and estimate choices for a project without changing Plane.
Inputs: project_id. Returns: object
get_work_item_relationsA
Read one work item's dependency, duplicate, schedule, and related-work relationships.
Inputs: work_item_id (required), project_id. Returns: object
list_configured_plane_workspacesA
List locally configured Plane workspaces and the project selected in each, without contacting Plane.
Inputs: no named parameters. Returns: object
list_plane_projectsA
List projects in the active Plane workspace so one can be selected without changing Plane.
Inputs: no named parameters. Returns: object
list_work_itemsC
List project work items with cross-edition local filters and bounded pagination.
Inputs: query, offset, cycle_ids, label_ids, +11 more. Returns: object
plan_git_backfillA
Read local git history and propose Plane work items for it, without changing Plane.
Commits are grouped by module_map path prefix (for example {"apps/web/": "Web"}), then by conventional-commit scope, then by top-level folder, and split when work pauses longer than max_gap_days. Start and target dates come from the commit dates, moved onto the project's business days. Review the items with the user, then call apply_git_backfill with the returned plan_id.
Inputs: since (required), repo_path (required), until, author, +4 more. Returns: object
save_project_workflow_profileB
Preview or save a project-bound workflow-profile override locally. Set confirm=true to write the profile file.
Inputs: profile (required), confirm, project_id, replace_existing. Returns: object
start_standard_work_itemB
Preview or start planned work using the project's configured started state and the actual start date.
Inputs: work_item_id (required), dry_run, state_id, project_id, +2 more. Returns: object
update_standard_work_itemC
Preview or update a specific Plane work item while preserving labels and description unless a structured replacement is requested.
Inputs: work_item_id (required), scope, context, dry_run, +22 more. Returns: object
upload_work_item_attachmentB
Preview or upload one local file as a Plane attachment. Set confirm=true to upload the file.
Inputs: confirm, file_path, project_id, max_size_mb, +2 more. Returns: object
validate_workflow_profileB
Validate a reusable workflow-profile override and return a representative title preview without saving it.
Inputs: profile (required). Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 33 tools
The core workflow actions (create/start/complete/cancel/update work items) and the git-backfill, reporting, relation, evidence, and attachment tools are clearly distinct. Several read-only status/context tools overlap around workspace, project, profile, setup, and diagnostic information, so an agent may need to inspect descriptions carefully to choose the right getter.
Tool names consistently use snake_case with a verb-first pattern such as get_, list_, create_, apply_, validate_, and add_. The few longer names remain predictable and readable, with no significant mixing of naming conventions.
At 33 tools, the set is heavy for a single MCP server even though the domain is broad. Many setup/status/profile helpers could likely be consolidated or grouped, making discovery and selection more burdensome than necessary.
The server covers workspace/project activation, workflow profiles, standardization planning, work-item lifecycle operations, relations, evidence links, attachments, reporting, auditing, and git backfill. Minor gaps remain around deletion/detachment operations and standalone comment management, but core workflows are well covered.
Maintenance
Related MCP Connectors
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for generating rough-draft project plans from natural-language prompts.
Related MCP Servers
- AlicenseCqualityDmaintenanceMCP server for Plane integration, enabling AI agents to interact with Plane APIs and services for project management tasks.100MIT
- FlicenseNot gradedqualityDmaintenanceMCP server and automation watcher for Plane project management, enabling Claude to interact with Plane via 40+ tools and auto-trigger on labeled issues.-
- AlicenseBqualityDmaintenanceAn MCP server that enables LLMs to fully manage a Plane workspace, including projects, work items, states, labels, cycles, modules, comments, and members.38MIT
- AlicenseBqualityAmaintenanceUnofficial MCP server for self-hosted Plane Community Edition, enabling project management tasks like managing work items, cycles, modules, and initiatives via natural language.1002MIT