The application infrastructure your agents reach for.
Connect smplkit’s MCP server once, and your coding agent — Claude, Cursor, Copilot, and more — can provision and operate every primitive in plain language: create a feature flag, set a config value, schedule a job, query the audit log, dial logging up to DEBUG. No dashboards to learn, no API to wire by hand. You stay the owner — every change runs under your account and shows up in your console.
“Schedule my cleanup endpoint to run nightly at 2am, and run it once to prove it works.”
Creates the job, fires it on demand, and shows you the real 200 OK — without leaving your editor.
And that’s one primitive. The same conversation works across all five:
- “Create a boolean flag
new-checkout, off by default, then turn it on in prod only for enterprise users.” - “List my environments, then set
database.hosttodb-staging.internalfor staging.” - “Raise the
sqlalchemy.enginelogger to DEBUG in production while I debug, then reset it.”
The agent makes the change and proves it worked — it can fire a job on demand or dry-run a SIEM forwarder before saving — then answers follow-up questions straight from the live state.
One MCP server, the whole platform.
Everything above is powered by a single Model
Context Protocol server, hosted at mcp.smplkit.com. MCP is the open standard coding agents
already speak, so smplkit works with Claude, Cursor, Copilot, and whatever agent ships next — no smplkit
plugin to install, no per-tool integration to maintain.
Cursor, Claude Desktop, and any other MCP client just point at the same URL from their config — and in Claude Code, connecting is one command:
claude mcp add --transport http smplkit https://mcp.smplkit.com/api/mcpThe first time your agent connects, your browser opens for a one-time sign-in — Google or Microsoft — and that’s it. There’s no key to mint, copy, or rotate. For CI and headless clients, authenticate with a scoped API key as a bearer token instead.
What your agent can operate
- Feature flags Create flags and set per-environment values, kill switches, and targeting rules.
- Config Create keyed, typed config and set values per environment.
- Logging Raise or reset runtime log levels per environment.
- Audit Search the audit log and manage SIEM forwarders.
- Scheduled jobs Create, run, and monitor scheduled HTTP jobs with full response capture.
- Environments List the account’s environments — the valid targets for every per-environment change.
The agent is increasingly the user. You remain the owner.
Every change is attributed
Changes run under your account — attributed to the signed-in user or to the API key — and appear in your console right next to the changes you make by hand.
Least privilege when you want it
With OAuth, the agent acts with your role. To narrow that, connect it with a scoped API key instead —
a VIEWER key lets an agent read state but never change it.
Nothing stored server-side
The MCP server is stateless and holds no credentials. Tokens are validated and used per request — never cached, never logged.
Five products, one server, one sign-in.