Stately MCP
Capped at 90: 1 of 2 real tool calls passed in our test; the tests add up to 93. Why scores are capped
Test points add up to 93; the score is capped at 90 because not every real tool call passed. Not a quality or security rating. Why scores are capped
Test points before the cap (93/100)
- Install20/20
- Handshake30/30
- Maintenance25/25
- Docs15/15
- License3/10
Tested npm @statelyai/mcp 0.28.0 (SEE LICENSE IN LICENSE; no source repo linked in package metadata). Installed pinned with scripts off (npm --ignore-scripts) and launched: initialize and tools/list over stdio succeeded with no credentials and listed 22 tools (22 annotated by the server: 19 read-only, 0 destructive). Real calls: 1 of 2 real function calls passed. Calls need a Stately API key (validate_machine answered 'Missing authentication').
Tool-call test (2026-11-13 UTC, direct MCP stdio client, no LLM): partial. 1 of 2 real function calls passed. Calls need a Stately API key (validate_machine answered 'Missing authentication'). Evidence: local test run, Nov 13, 2026 UTC.
Transport (tested 2026-11-13 UTC): stdio
Token cost (estimate): approx. 11173 tokens, cl100k estimate (22 tools, measured 2026-11-13 UTC, version 0.28.0). This is an estimate from our testing of the tool names, descriptions and input schemas, counted with cl100k_base.
Permissions: 8 read, 3 write (state-changing), 11 unclassified · Auth: API key · Rate limit: unknown · Pricing unit: unknown
Auth: API key. Rate limit: unknown. Pricing unit: unknown.
apply_machine_patches— write (state-changing)check_machine— unclassifiedconvert_machine— unclassifieddiff_machines— readexport_machine_image— unclassifiedextract_machines_from_code— unclassifiedgenerate_graph_paths— unclassifiedget_machine— readget_stately_machine— readget_stately_project— readlist_machines— readlist_stately_project_machines— readlist_stately_projects— readread_visualization_workspace— readreport_visualization_rendered— unclassifiedreview_machine_implementation— unclassifiedsimulate_machine— unclassifiedupdate_visualization_workspace— write (state-changing)validate_machine— unclassifiedvisualize_machine— unclassifiedvisualize_machines— write (state-changing)visualize_stately_machine— unclassified
listing-flags calls_need_key or auth_kind
Score 90/100 · Tools 22 · Token cost approx. 11,173 tokens · Compare · Last tested 2026-11-13 UTC · Version tested 0.28.0 · Package @statelyai/mcp
Summary: Stately MCP completed an MCP handshake without credentials and exposed 22 tools.
Evidence summary
- Handshake
- pass
- Tools listed
- 22
- Context cost
- approx. 11173 tokens, cl100k estimate (22 tools, measured 2026-11-13 UTC, version 0.28.0; an estimate) ranking · add up your servers' context cost
- API key
- Connects without a key; tool calls need one
- Tested
- 2026-11-13 UTC
Test results
| Test | Result | Points | Evidence |
|---|---|---|---|
| install | pass | 20 | npm install @statelyai/mcp (Node 20, Linux) succeeded in 1s. |
| handshake | pass | 30 | MCP initialize + tools/list over stdio succeeded with no credentials; 22 tools listed in 5.5s. |
| maintenance | pass | 25 | Latest release published 1 days before test. |
| docs | pass | 15 | README over 3,000 chars; install instructions present; mcpServers config example present. Check of the package's own README (README shipped in npm package). |
| license | partial | 3 | License not machine-detected as OSI (SEE LICENSE IN LICENSE). |
Local-model readiness
Token cost (estimate): approx. 11173 tokens, cl100k estimate for 22 tools · about 136% of an 8k context. This is an estimate from our testing (counted with cl100k_base).
Small-model friendly (schema heuristic, not a model result): unknown.
Not tested
Small-model task runs: Not yet tested.
What we tested and what we didn't
Tested on 2026-11-13 UTC
- Install from the registry
- We connected with no credentials and asked for its tool list: it listed 22 tools (Method details: MCP
initializethentools/list) - Maintenance signals (last push, archived/deprecated status)
- Docs (README length, install and config example)
- License (OSI-approved or not)
- Token cost (estimate)
Not tested
- Security: not audited (no code review, no permission or data-flow analysis)
- Real tasks (no functional task runs yet)
- Behaviour with real credentials or a real account
- Performance, rate limits or load
- Correctness of results beyond the listed tasks
Methodology: how we test. A high score means it installs and responds, not that giving it access to your data carries no risk.
Known security issues
Not yet checked.
Monthly roundup: MCP security issues, October 2026.
Tool list changes
Baseline recorded from the 2026-11-13 UTC test run; changes appear after the next test.
Snapshot hash (fingerprint of the tool list)
65e5313b5ea2: SHA-256 of the canonical tool names, descriptions and input schemas. If it changes, the tool list changed.
Raw test record (JSON)
{
"name": "@statelyai/mcp",
"version": "0.28.0",
"registry": "npm",
"status": "working",
"tested": "2026-11-13",
"tools_count": 22,
"deprecated": false,
"archived": null,
"tests": {
"install": [
"pass",
20,
"`npm install @statelyai/mcp` (Node 20, Linux) succeeded in 1s."
],
"handshake": [
"pass",
30,
"MCP `initialize` + `tools/list` over stdio succeeded with no credentials; 22 tools listed in 5.5s."
],
"maintenance": [
"pass",
25,
"Latest release published 1 days before test."
],
"docs": [
"pass",
15,
"README over 3,000 chars; install instructions present; `mcpServers` config example present. Check of the package's own README (README shipped in npm package)."
],
"license": [
"partial",
3,
"License not machine-detected as OSI (SEE LICENSE IN LICENSE)."
]
}
}Install
npx -y @statelyai/mcp
Pin the version we tested
npx -y @statelyai/mcp@0.28.0
tested 2026-11-13 UTC. Newer versions may differ and were not tested by us.
{
"mcpServers": {
"stately-mcp": {
"command": "npx",
"args": [
"-y",
"@statelyai/mcp@0.28.0"
]
}
}
}Tools listed
list_machines, get_machine, visualize_machines, read_visualization_workspace, update_visualization_workspace, validate_machine, convert_machine, diff_machines, apply_machine_patches, review_machine_implementation, generate_graph_paths, simulate_machine, check_machine, export_machine_image, visualize_machine, extract_machines_from_code, report_visualization_rendered, visualize_stately_machine, list_stately_projects, get_stately_project, list_stately_project_machines, get_stately_machine
Quick config
{
"mcpServers": {
"stately-mcp": {
"command": "npx",
"args": [
"-y",
"@statelyai/mcp"
]
}
}
}Add to config Opens the config generator with this server selected.
Setup
- Install. Run
npm install @statelyai/mcp(Node 20). This is the exact command we ran on 2026-11-13 UTC. - Configure. Add the mcpServers entry below to your MCP client config.
- Verify. Restart the client and check that tools/list returns tools (22 in our test).
{
"mcpServers": {
"stately-mcp": {
"command": "npx",
"args": [
"-y",
"@statelyai/mcp"
]
}
}
}Test history
- 2026-11-13 UTC: New
Related: Best dev MCP servers
FAQ
Is Stately MCP deprecated?
No. @statelyai/mcp was not deprecated or archived when we last tested it on 2026-11-13 UTC.
What replaces Stately MCP?
Nothing; Stately MCP is not deprecated as of 2026-11-13 UTC.
Does Stately MCP need an API key?
It connects and lists 22 tools without a key, but its tool calls need one.
Embed the badge
Every tested server gets a free embeddable badge. It shows the current score and test date and expires after 90 days. Testing is free; payment never changes a score.