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., "@AgilePlace MCP Server Bulk create 30 cards on my 'Sprint 12' board"
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.
AgilePlace demo-provisioning MCP server. Creates and shapes demo environments (boards, bulk cards, hierarchy) from Claude Desktop or Cursor. stdio transport, single-user.
Module | Role |
| Env-driven config + authed HTTP client (Bearer). |
| Board-list warm on first use; lazy per-board card types; label → ID resolution. |
| Throttles off |
| Bulk card creation. |
| stdio MCP entrypoint. |
Related MCP server: Trello MCP Server
Sequential, self-throttling writes.
Max ~40 cards per call; larger batches rejected (chunk and retry).
Per-card clientKey correlation, echoed on results.
Error handling: retry (429 / 5xx / network), fail-card-continue (400 / 422 / 404), abort-batch (401 / 403), circuit breaker on repeated identical failures.
Partial success returns isError: false; cards in created[] should not be resubmitted.
Set credentials in the host config env block (claude_desktop_config.json or Cursor mcp.json). Not in source.
{
"mcpServers": {
"agileplace-v3": {
"command": "node",
"args": ["/absolute/path/to/src/server.mjs"],
"env": {
"AGILEPLACE_DEFAULT_URL": "https://<your-org>.leankit.com/io",
"AGILEPLACE_DEFAULT_TOKEN": "<token>"
}
}
}
}Run standalone:
node src/server.mjscreateCards — implemented.
updateCards / findCards / createBoard — not yet implemented.
Deferred: multi-environment routing, lane name resolution, OKR tools, whole-call idempotency, tasks extension.
Bulk-creates cards on one board. Use for new cards, not updates. Returns created[] with clientKey + the real id for a later connect pass; failed[] carries clientKey + title + reason. Pass type as a label ("Epic"); the tool resolves it. header is cosmetic and is not used to connect cards. One board per call. Max 40 cards per call — chunk larger sets and concatenate created[] before connecting. Partial success is normal: isError: false with some cards in failed[]. Cards in created[] already exist — NEVER resubmit those clientKeys. To complete a partial batch, resubmit only the failed[] (and any never-sent) cards. This applies to every partial result, not just retryAfter rate-limit cases. If retryAfter is set, wait that many seconds, then resubmit only the never-sent cards (not created[]). Never invent board, type, or lane ids. Attach to an existing parent with parentCards; otherwise create all cards first, then connect using the returned ids.
| Name | Required | Description | Default |
|---|---|---|---|
| cards | Yes | Cards to create. Max 40 per call; chunk larger sets. | |
| boardId | Yes | Target board title or id. One board per call. |
| Name | Required | Description |
|---|---|---|
| failed | Yes | |
| created | Yes | |
| summary | Yes | |
| retryAfter | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Goes far beyond annotations. The description discloses partial success semantics (isError: false with failed[]), idempotency implications (created[] cards already exist), the tool's resolution of type labels, the cosmetic nature of header, and explicit prohibitions like 'Never invent board, type, or lane ids.' These behaviors are not inferable from annotations alone.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is long and dense, but each sentence provides essential operational detail for handling partial failures, rate limits, and parent-card connections. It is front-loaded with the core purpose and then systematically covers edge cases. Slightly more structured formatting (bullets) could improve readability, but nothing is wasteful.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (bulk creation, partial success, retryAfter, parent linkage) and the presence of an output schema, the description is exceptionally complete. It explains return values, error handling, chunking, idempotency rules, and integration steps, leaving no significant ambiguity for an AI agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Although the schema already describes all parameters (100% coverage), the description enriches meaning by clarifying usage: 'type' is resolved to typeId, 'header' is cosmetic and not used for connections, 'clientKey' is echoed in responses, and 'boardId' accepts title or id. These are practical semantics that go beyond mere field descriptions.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description opens with 'Bulk-creates cards on one board,' a specific verb+resource statement that clearly identifies the tool's function. It also explicitly distinguishes from updates ('Use for new cards, not updates'), which helps differentiate even in the absence of sibling tools.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides explicit guidance on when to use the tool ('Use for new cards, not updates') and how to handle failures ('resubmit only the failed[] cards'). It covers chunking (>40), partial success, retryAfter logic, and never resubmitting created[] clientKeys, giving clear directives for safe and correct usage.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v3.0.0createCardsScored across 1 tool
Only one tool exists, so there is no possibility of overlap or misselection. Its purpose is explicit: bulk-create cards on a board.
The single tool uses a clear verb_noun camelCase name, createCards, but there is no broader naming pattern to evaluate consistency across a set.
For an AgilePlace server, one tool covering only card creation is too narrow; agents will likely need other board, lane, and card lifecycle operations. The tool itself is substantial, but the server is under-scoped.
Only creation is exposed; there are no read, update, delete, search, or connect operations for cards or boards. Bulk creation is handled robustly, but most AgilePlace workflows hit dead ends.
Create and drive KanbanThing kanban boards. No account, no API key, board link required.
Read and update Markboard kanban boards: cards, checklists, labels, comments and files.
ClickUp MCP — wraps the ClickUp REST API v2 (BYO API key)
Read and write Mission Control state via MCP — projects, tasks, subtasks, templates, status updates.
