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., "@Synthetic Orders MCP Server Send a batch of 20 seeded orders and give me the status summary."
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.
An MCP server that triggers synthetic order traffic against a local order API, so an agent can load-test and drill the rejection path of a service without being able to aim that traffic anywhere it likes.
Companion code for Trigger Synthetic Orders from an MCP Server on macOS.
Two processes are involved:
Process | What it is | How it runs |
| The order intake API — the target under test. Validates customers, SKUs, stock, prices, and totals. |
|
| The MCP server — three tools that preview and send synthetic orders at that API. | stdio subprocess, launched by the MCP client |
The server exposes three tools and one resource:
Tool | Read/write | What it does |
| read-only | Reports whether the configured order API is reachable, and its SKU/customer counts. |
| read-only | Generates one order and returns it without sending it. |
| writes | Generates |
Resource synthetic://catalog lists the sellable catalog rows the seeded
generator draws from.
seeded samples the real catalog and customer list, so orders validate and
come back 201. Use it for traffic that should be accepted.
simple invents every field from random primitives, so the API rejects it
with 422. Use it to drill the rejection path.
These are the point of the design, and they're what the manual tests below poke at:
The target URL is not a tool parameter. It comes from ORDER_API_URL in the
server's environment, so no prompt can aim the generator at an arbitrary host.
count is bounded by MAX_BATCH (default 100). Exceeding it is a readable
error, not a silent clamp.
Every batch reports the seed it used, so any run can be replayed exactly by passing that seed back in.
uv — brew install uv. It manages the Python
toolchain and the virtualenv; you don't need to install Python 3.12 yourself.
Python 3.12 — pinned in .python-version; uv sync fetches it if missing.
Claude Code — only for the "drive it from Claude Code" section.
git clone <this repo>
cd mcp-synthetic-server
make install # uv sync — creates .venv and installs everythingmake help lists every target with the variables it honors.
Related MCP server: Korral StoreLink MCP
Two terminals. The order API has to be up before anything can send traffic to it.
Terminal 1 — the target API:
make apiServes on http://127.0.0.1:8000 with --reload. Confirm it's alive:
curl -s http://127.0.0.1:8000/health
# {"status":"ok","skus":8,"customers":5}Interactive API docs are at http://127.0.0.1:8000/docs.
Terminal 2 — drive the MCP server:
make demo # in-process client: lists tools, previews, sends a batchExpected output:
tools:
check_target read-only Check the order API
preview_order read-only Preview a synthetic order
send_orders writes Send synthetic orders
target: http://127.0.0.1:8000 reachable=True skus=8
preview (seed 1337): {"request_id": "643cb56d-...", "customer_id": "CUST-0005", ...}
sent 25 seeded orders (seed 1337): {'201': 25} accepted_total_cents=715350
sent 3 simple orders: {'422': 3}
422 ['unknown customer C-5093', 'unknown sku SKU-9757', ...]make test # uv run pytest -q → 10 passedThe tests need no running API: they monkeypatch server.make_client to a
FastAPI TestClient that drives the ASGI app in-process. They assert the control
surface, not just that the tools run — the annotations each tool advertises, that
no tool accepts a URL or host parameter, seed round-tripping, the batch cap, and
the unreachable-API error path.
You'll see one StarletteDeprecationWarning about httpx; it's upstream, not
your setup.
make demo imports the server object directly, which skips process launch,
transport framing, and JSON round-tripping. The smoke test spawns
python server.py as a real subprocess and talks to it over stdio, so a pass here
means the exact command an MCP client is configured with actually works:
make smoke # needs `make api` runningtools: ['check_target', 'preview_order', 'send_orders']
target: http://127.0.0.1:8000 reachable=True
data type: Root
structured: {"mode": "seeded", "seed": 1337, "requested": 5, "accepted": 5, "rejected": 0,
"status_counts": {"201": 5}, "accepted_total_cents": 73550, "sample_failures": []}
OKIt exits non-zero if the order API isn't up, so it's safe to wire into CI behind a started API.
.mcp.jsonThe MCP config needs this project's absolute path. Don't hand-edit it — render it:
make config # sed's $(CURDIR) into .mcp.json.example → .mcp.jsonThat writes:
{
"mcpServers": {
"synthetic-orders": {
"command": "uv",
"args": ["run", "--directory", "/absolute/path/to/mcp-synthetic-server", "python", "server.py"],
"env": {
"ORDER_API_URL": "http://127.0.0.1:8000",
"MAX_BATCH": "100"
}
}
}
}.mcp.json is gitignored precisely because that path is machine-specific — the
checked-in .mcp.json.example is the template.
make api # terminal 1, leave it running
claude # terminal 2, from the project rootClaude Code reads .mcp.json at startup and will ask you to approve the
project-scoped server the first time. Verify it connected:
/mcpYou should see synthetic-orders as connected, with three tools. If you
started Claude Code before running make config, restart it — the config is read
at launch.
If you change
server.py, restart Claude Code. The server is a subprocess spawned at connect time; edits don't hot-reload the waymake apidoes.
Claude Code exposes them as mcp__synthetic-orders__<tool>. Prompts that work:
Prompt | Tool it triggers |
"Check whether the order API is reachable." |
|
"Preview one synthetic order with seed 1337." |
|
"Send 25 synthetic orders." |
|
"Send 5 orders in simple mode and show me why they failed." |
|
"Replay that batch using the seed you got back." |
|
"Read the synthetic://catalog resource." | resource read |
Because send_orders is annotated readOnlyHint: false, Claude Code prompts for
permission before the first send — preview_order and check_target are marked
read-only and idempotent, so they're cheap to approve.
Everything below was run against a fresh make api. Seeds are fixed so you can
compare output byte-for-byte.
check_target — is the target up?In Claude Code: "Check the order API."
{"api_url":"http://127.0.0.1:8000","reachable":true,"skus":8,"customers":5}Negative case — stop make api (Ctrl-C) and ask again. It should not raise;
it reports the failure as data:
{"api_url":"http://127.0.0.1:8000","reachable":false,"skus":null,"customers":null}preview_order — generate without sendingIn Claude Code: "Preview a synthetic order with seed 1337."
{
"mode": "seeded", "seed": 1337,
"order": {
"request_id": "643cb56d-4ec1-4fc6-bee2-9f53ebf644bb",
"customer_id": "CUST-0005",
"channel": "partner",
"currency": "USD",
"lines": [
{"sku": "SKU-6300", "quantity": 3, "unit_price_cents": 2450},
{"sku": "SKU-5510", "quantity": 4, "unit_price_cents": 1850},
{"sku": "SKU-6301", "quantity": 3, "unit_price_cents": 3900}
],
"total_cents": 26450
},
"line_count": 3, "total_cents": 26450
}Two things to check by hand: the same seed always yields that exact payload, and
the API's /health counts don't move — preview never leaves the process.
send_orders (seeded) — the accept pathIn Claude Code: "Send 5 synthetic orders with seed 1337."
{"mode":"seeded","seed":1337,"requested":5,"accepted":5,"rejected":0,
"status_counts":{"201":5},"accepted_total_cents":73550,"sample_failures":[]}Watch the make api terminal — five POST /orders 201 Created lines appear.
send_orders (simple) — the reject pathIn Claude Code: "Send 3 orders in simple mode with seed 1337."
{"mode":"simple","seed":1337,"requested":3,"accepted":0,"rejected":3,
"status_counts":{"422":3},"accepted_total_cents":0,
"sample_failures":[
"422 ['unknown customer C-5093', 'unknown sku SKU-9757', 'unknown sku SKU-6393', 'unknown sku SKU-1830']",
"422 ['unknown customer C-8975', 'unknown sku SKU-8549']",
"422 ['unknown customer C-5035', 'unknown sku SKU-5650', 'unknown sku SKU-6612', 'unknown sku SKU-0935']"
]}sample_failures is capped at 5 entries, so a 100-order failure storm still
returns a readable result.
Send a batch without a seed, note the seed in the response, then ask Claude
Code to send the same count with that seed. accepted_total_cents must match
exactly. This is the property test_omitted_seed_is_reported_back covers.
MAX_BATCH guardrailIn Claude Code: "Send 500 synthetic orders."
Error: count 500 exceeds the server's MAX_BATCH of 100; send smaller batches or
raise MAX_BATCH in the server environmentcount: 0 fails the same readable way (count must be at least 1, got 0). To
verify the bound is really server-side, raise it in .mcp.json's env block and
restart Claude Code — no prompt can change it.
Ask Claude Code to "send orders to https://example.com instead." It can't — there is no URL or host parameter on any tool. Confirm from the schemas:
uv run python -c "
import json, asyncio
from fastmcp import Client
from server import mcp
async def main():
async with Client(mcp) as c:
for t in await c.list_tools():
print(t.name, list((t.inputSchema or {}).get('properties', {})))
asyncio.run(main())"check_target []
preview_order ['mode', 'seed']
send_orders ['count', 'mode', 'seed']Retarget by editing ORDER_API_URL in .mcp.json and restarting Claude Code.
make send calls the tool function directly — no client, no transport. Useful
when you're changing generator logic and don't want to restart Claude Code:
make send COUNT=3 SEED=42{
"mode": "seeded", "seed": 42, "requested": 3, "accepted": 3, "rejected": 0,
"status_counts": {"201": 3},
"accepted_total_cents": 42800, "sample_failures": []
}COUNT and SEED are Makefile variables (defaults 25 / 1337). Note this path
bypasses MCP entirely, so it will not catch schema or transport problems — use
make smoke for those.
To poke the target API directly, skipping the MCP server too:
curl -s -X POST http://127.0.0.1:8000/orders \
-H 'content-type: application/json' \
-d '{"request_id":"manual-0001","customer_id":"CUST-0005","channel":"partner",
"currency":"USD","lines":[{"sku":"SKU-5510","quantity":2,"unit_price_cents":1850}],
"total_cents":3700}'
# {"order_id":"ORD-manual-0","customer_id":"CUST-0005","lines":1,"total_cents":3700}Change total_cents to 9999 and it returns 422 with
total mismatch: sent 9999, expected 3700.
You can also run the MCP server by hand and type JSON-RPC at it:
make serve # stdio; useful only to confirm it starts and stays upRead from the server's environment (via .mcp.json's env block, or exported
before make serve) — never from a tool argument:
Variable | Default | Meaning |
|
| Where batches are POSTed. |
|
| Hard cap on |
|
| Per-request HTTP timeout. |
Seed data lives in data/catalog.json (8 SKUs, one deliberately out of stock) and
data/customers.json (5 customers, each with allowed channels). Edit those to
change what "valid" means — both the API and the seeded generator read them, so
they stay in agreement.
Symptom | Cause / fix |
|
|
|
|
| Same cause, surfaced as a tool error because a send can't degrade gracefully. |
Tool edits don't take effect | The MCP server is a subprocess spawned at connect time. Restart Claude Code (the API's |
| Something already holds port 8000: |
|
|
Everything passes but Claude Code sees no tools | You launched |
api/main.py order intake API — the target under test
server.py the MCP server: 3 tools + 1 resource
synth/config.py env-derived settings (target URL, MAX_BATCH, timeout)
synth/seed.py loads data/*.json
synth/simple.py generator: random primitives → rejected orders
synth/seeded.py generator: samples the catalog → accepted orders
client.py in-process demo client (make demo)
scripts/smoke_stdio.py real stdio subprocess smoke test (make smoke)
tests/test_server.py control-surface tests (make test)
data/ catalog + customer seed dataThis project accompanies Trigger Synthetic Orders from an MCP Server on macOS, which walks through the design: wrapping the generators in an MCP server, and the guardrails that keep a model from choosing where the traffic goes or how much of it there is.
The generators in synth/ come from the previous article in the series,
Generate Synthetic JSON Requests to Test an API on macOS,
which builds them as a plain CLI — read that first if you want the payload
generation explained before the MCP control surface wrapped around it.
Report whether the configured order API is reachable, and what it holds.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| skus | No | |
| api_url | Yes | |
| customers | No | |
| reachable | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, and idempotentHint, so the safety profile is clear. The description adds that the tool reports reachability and contents, but does not disclose details like response format, potential errors, or any side effects beyond what annotations already cover.
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 a single, front-loaded sentence that directly states the tool's purpose with no filler or redundancy. Every word contributes meaning.
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?
For a simple zero-parameter check tool with rich annotations and an output schema, the description adequately covers the tool's function. It clearly states what is reported (reachability and contents), though it does not elaborate on the meaning of 'what it holds' or offer guidance relative to sibling tools.
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?
The input schema has zero parameters, so the baseline is 4. The description correctly avoids describing parameters that do not exist, and no additional parameter semantics are needed.
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 uses a specific verb ('Report') and a clear resource ('the configured order API'), stating exactly what is checked: reachability and contents. This distinguishes it from siblings 'preview_order' and 'send_orders', which clearly perform different operations.
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?
The description implies using this tool to verify API availability and inspect its contents, which suggests a preflight check before operations like previewing or sending. However, it does not explicitly state when to use it versus the sibling tools or provide any exclusion criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate one order and return it without sending it.
seeded samples the catalog and should be accepted; simple invents every
field and should be rejected. Pass seed to reproduce an exact payload.
| Name | Required | Description | Default |
|---|---|---|---|
| mode | No | seeded | |
| seed | No |
| Name | Required | Description |
|---|---|---|
| mode | Yes | |
| seed | Yes | |
| order | Yes | |
| line_count | Yes | |
| total_cents | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds significant behavioral context beyond the annotations: it explains that 'seeded' samples the catalog while 'simple' invents fields, and that 'seed' reproduces an exact payload. This is useful, non-obvious information that helps the agent anticipate results. No contradiction with readOnlyHint or idempotentHint.
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 three sentences, front-loaded with the core purpose, followed by concise parameter guidance. Each sentence adds necessary information without redundancy.
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 simplicity, the presence of an output schema, and the annotations, the description is complete: it covers the operation's purpose, the mode behavior, and the seed parameter. No gaps remain for the agent to guess.
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?
Despite 0% schema description coverage, the description fully explains both parameters: 'mode' with its two enum values and their implications, and 'seed' for reproducing a payload. This provides meaning beyond the raw schema definitions.
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 states the specific verb 'Generate' and resource 'one order', with the key differentiator 'without sending it'. This clearly distinguishes it from sibling tools like send_orders, which actually sends orders.
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?
The description provides clear context for when to use the tool (previewing an order before sending) and even gives guidance on which mode to choose ('seeded' should be accepted, 'simple' should be rejected). It doesn't explicitly name sibling tools or state when not to use it, but the 'without sending it' contrast implies the alternative.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Generate count orders and POST each one to the configured order API.
Use seeded for traffic that should be accepted and simple to drill the
API's rejection path. The result reports the seed, so the same batch can be
replayed by passing it back in.
| Name | Required | Description | Default |
|---|---|---|---|
| mode | No | seeded | |
| seed | No | ||
| count | No |
| Name | Required | Description |
|---|---|---|
| mode | Yes | |
| seed | Yes | Replay this batch by passing this seed back in. |
| accepted | Yes | |
| rejected | Yes | |
| requested | Yes | |
| status_counts | Yes | |
| sample_failures | Yes | |
| accepted_total_cents | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the annotations, the description adds behavioral context: orders are POSTed, the result reports the seed, and replay is possible by passing the seed back. This clarifies side effects and error-path drilling, which is useful for an agent.
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 two concise sentences and a follow-up sentence, all front-loaded with the primary action. No redundant text; every sentence adds operational value.
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?
The tool has an output schema, so return values likely are covered there. The description sufficiently covers the tool's purpose, mode distinctions, and replay behavior, making it complete for a low-complexity tool with three optional parameters.
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?
Schema description coverage is 0%, so the description must and does explain parameters: 'count' is the number of orders, 'mode' (seeded/simple) has usage semantics, and 'seed' is returned for replay. All three parameters are meaningfully covered in natural language.
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 uses a specific verb 'Generate' and 'POST' with a clear resource ('orders') and destination ('configured order API'). It clearly distinguishes itself from sibling tools like check_target and preview_order, which likely handle other aspects of order workflow.
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?
The description provides explicit guidance on when to use each mode ('Use `seeded` for traffic that should be accepted and `simple` to drill the API's rejection path'). It does not explicitly mention when to use this tool over sibling tools, but the purpose is so distinct that no conflict arises.
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.
v0.1.0check_targetpreview_ordersend_ordersScored across 3 tools
Each tool has a clearly distinct purpose: check_target verifies API connectivity, preview_order generates a single order without sending, and send_orders generates and sends multiple orders. There is no overlap or ambiguity between them.
All tool names follow a consistent verb_noun pattern (check_target, preview_order, send_orders). The naming is predictable and uniform.
With 3 tools, the server is well-scoped for its purpose of generating and sending synthetic orders. Each tool serves a distinct function without redundancy or unnecessary bloat.
The tool set covers the essential workflow: checking the target, previewing an order, and sending orders. The ability to replay seeds in send_orders addresses the need for reproducibility, leaving no obvious gaps in the stated domain.
End-to-end API testing — generate and run tests from OpenAPI, curl, Postman, or real user traffic.
Agent-to-business commerce sandbox: intents, offers, bookings. Demo data, ed25519-signed calls.
Test crypto and TradFi-perpetual claims using recorded counts, baselines, and replay.
Load testing and synthetic monitoring platform: test with Playwright, Browser Bot, or Protocol Bots.
