Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@adhd-mcp estimate my context and readiness for this task"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
ADHD-friendly Cursor agent kit: on-demand Q&A, concise replies with status footer, context stats via MCP.
cd your-project
npx adhd-mcp initReload MCP in Cursor → new Agent chat.
Related MCP server: fcop-mcp
init writesPath | Role |
| Output baseline + Q&A gate + coding ladder |
| Q&A clarification workflow |
| Short reply + footer format |
| Distinctive UI (anti–AI-slop) |
| Vercel Web Interface Guidelines audit |
| Merges |
| Q&A session logs |
estimate_context — context % (reported or ~heuristic)
format_status_footer — 📊 理解 X% │ 进度 Y% │ ctx ~Z%
check_readiness — block code until understanding ≥95%
verify_against_requirement — after edits: user wording vs coding ladder
qa_session_start / qa_session_append — session files
✅ Rule + Skills + MCP + init
❌ GitHub search (v2) · hitl integration
npm install
npm run smoke
npm run local-test
node dist/index.js init --cwd /path/to/project --force --localAfter publish, users run npx adhd-mcp init (uses npm; no --local).
MIT
Returns whether understanding ≥95% — code edits forbidden when ready=false.
Inputs: understanding_pct (required).
Estimate or accept reported context window usage %. Call before format_status_footer when IDE ring % is unknown.
Inputs: turn_count, rules_chars, conversation_chars, mcp_overhead_chars, +1 more.
Append one Q&A entry to session file. Ask only ONE question per chat turn.
Inputs: kind (required), content (required), session_path (required), understanding_pct.
Create adhd/sessions/YYYY-MM-DD-slug.md when starting Q&A for a task.
Inputs: task_summary (required), slug, workspace.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 5 tools
Each tool has a clearly distinct purpose: context estimation, footer formatting, readiness check, session creation, and session appending. No two tools could be confused for one another.
All tool names follow a consistent verb_noun pattern (estimate_context, format_status_footer, check_readiness, qa_session_start, qa_session_append). Perfectly predictable.
Five tools is well-scoped for this niche server. Each tool earns its place and covers a specific step in the workflow without redundancy.
The surface fully covers the intended domain: context awareness, mandatory formatting, readiness gating, and Q&A session lifecycle. No obvious gaps for the stated purpose.
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
Agent-native notes, tasks, dev-docs, vaults, sync & handoffs. MCP + OpenAPI dual surface.
ADHD system of record for agents: tasks, goals, loops, calendar, focus stats.
Cross-tool persistent memory and context for AI assistants over MCP.
