From management-portal
Operate the management-portal MCP correctly — the freelancer/client workspace of clients, projects, briefs, proposals, phases, milestones, tasks, subtasks, a flow board, calendar, notes, boards, gigs, time, and team chat. Use whenever you read or write any management-portal / portal entity (create or update a client/project/brief/proposal/milestone/task, build or edit the flow board or a board, run board-first alignment, or break work into tasks + clusters + relations). Enforces the core loop, the three gates (read-after-write, completeness, task-breakdown), bottom-up completion, never-fabricate-ids, and board-first.
How this skill is triggered — by the user, by Claude, or both
Slash command
/management-portal:management-portalThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You operate a **real freelancer's workspace** over MCP tools. **Writes change real data.** You have the
You operate a real freelancer's workspace over MCP tools. Writes change real data. You have the
tools, not the database — the read tools (get_*/list_*) are your only source of truth and your only
way to verify. Ground every action in what the portal actually contains, then leave it complete and
correct. Never guess a field, an id, or a structure you could read.
READ → GAP → ALIGN(board-first) → BREAK DOWN → BUILD → TEST → VERIFY → DELIVER → UPDATE
reference.md §board-first.get_*/list_* and confirm the field
persisted. Trust the data effect, not the success string or a stale schema. (create_flow_connection
→ list_flow_connections; the edge may land even on a reported timeout.)Bottom-up completion — a parent is done only when every child is verified done:
list_subtasks(parent) → do + complete_task(child) → re-list → then complete the parent.
Never fabricate ids — every id comes from a list_*/get_* read or a create_* response. Never
invent, guess, pattern-match, abbreviate, or reuse-from-memory an id. No id? Read for it.
For any non-trivial new work, align on a board with diagrams BEFORE the brief/proposal/task tree:
create_board → blocks (callout/heading/text + mermaid + charts) → read_board to verify →
present and stop until the human aligns → only then build. Insert mermaid via insert_diagram /
create_board_block(type:"mermaid"), never raw mermaid pasted into a field.
Whatever intelligence is in the loop self-extracts — there is no separate extractor for the MCP. As an MCP agent you extract yourself: reason over read-tool output, then write clusters/relations/tasks via primitive tools. No DeepSeek (that's only for server-side app buttons like Generate Tasks). The in-app AI Chat / Thoth likewise self-extract.
Read reference.md (same folder) before you actually build, for: the full discipline summary, the
principle → exact-tool-sequence playbook (create project + brief + proposal, add phase + milestones,
cost/time, task breakdown + flow board, bottom-up completion), the write → read verification map (which
read tool confirms each write), and the board-first procedure with cluster/relation conventions.
Connection params: ../../mcp.config.md. Canon: ../../../DISCIPLINE.md and companions.
npx claudepluginhub waelfouda/management-portal-plugin --plugin management-portalProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.