Skills & Agents

Monday Morning ships with a curated library of Claude Code skills — the /mm: command set — and a roster of agents that do the heavy lifting behind those commands. Skills are managed per project: you can browse what’s installed, import new ones from a URL or a local file, and toggle them on or off for the project you’re in. There is no central registry to publish to or pull from; everything lives with your workspace.

Agents come in two flavors. Workflow agents drive the spec pipeline end-to-end. Specialist agents are surfaced as clickable chips in the terminal so you can hand a focused task to the right expert without writing a prompt from scratch.

Workflow agents

These agents run the spec lifecycle: shaping requirements, writing the spec, breaking it into a task list, implementing, and verifying. They are invoked by the /mm: skills that orchestrate the pipeline, and several run in their own sessions so stages can proceed in parallel.

Agent	What it does
spec-initializer	Scaffolds a new spec folder and its starting files
spec-shaper	Gathers requirements through targeted questions and visual analysis
spec-writer	Authors the spec from the shaped requirements
spec-verifier	Reviews the spec for completeness and consistency
tasks-list-creator	Breaks the spec into an implementable task list
implementer	Implements the tasks against the codebase
implementation-verifier	Checks the implementation against the spec and tasks
adversarial-verifier	Stress-tests the work, hunting for gaps and regressions
product-planner	Defines product vision and roadmap during planning

The flow is roughly: shape → write → verify spec → task list → implement → verify implementation, with the adversarial pass as an extra layer of scrutiny.

Specialist agents

Specialist agents are the mm-* agents surfaced as clickable chips in the terminal. Each is scoped to one discipline, so you can route a review, audit, or fix to the agent best suited for it.

Agent	What it does
mm-ui-ux-reviewer	Reviews UI and UX for usability and flow issues
mm-visual-designer	Handles visual design and styling work
mm-integration	Wires integrations end-to-end across the frontend/backend boundary
mm-a11y-auditor	Audits for accessibility issues
mm-debugger	Debugs and traces issues to root cause
mm-qa-engineer	Plans and runs QA and testing
mm-perf-optimizer	Profiles and optimizes performance
mm-security-auditor	Audits for security vulnerabilities

Because specialists run in their own sessions, they compose naturally with the parallel and fleet workflow — you can dispatch several at once and let them work independently while the main spec pipeline continues.

Managing skills

Skill management is per project. From the management UI you can:

• Browse the skills installed for the current project
• Import a skill from a URL or a local file
• Toggle skills on or off so each project only carries what it needs
• Organize skills per project, with no central registry to sync against

Because configuration is per project, two workspaces can carry entirely different skill sets, and turning a skill off in one project doesn’t affect another.

Related MCP tools

Several MCP tools back the skill and agent surfaces:

Tool	Purpose
mm_create_skill	Create a new skill
mm_import_skill	Import a skill from a URL or local file
mm_list_plugins	List installed plugins
mm_get_plugin_info	Inspect a specific plugin

Related

• Spec Retrieval (Prior Work) — how completed specs prime new ones in the same pipeline
• Your First Workflow — the /mm:spec lifecycle the workflow agents drive