PWA Today MCP Server
OfficialAllows running PWA Today runtime audits on a given URL, checking audit status, and retrieving audit results.
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., "@PWA Today MCP ServerRun a PWA Today audit on https://example.com and show me the results."
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.
PWA Today MCP Server
Run generic PWA checks and PWA Today runtime audits from an MCP-compatible coding agent.
Never paste your PWA Today client secret or access token into a chat. Store it in your MCP client's secret or environment configuration. The model does not need to see it.
Setup
Create or retrieve API credentials in the PWA Today console.
Register and verify the application hostname in the console if you will run hosted audits.
Install the package or configure your MCP client to run
pwa-today-mcp.Store
PWA_CLIENT_IDandPWA_CLIENT_SECRETin the server environment. Optionally setPWA_API_URL; it defaults tohttps://api.pwa.today.Restart or reconnect your MCP client.
Ask your agent to start an audit.
The server uses stdio only and exposes these tools:
check_pwastart_pwa_auditcreate_pwa_audit_configurationget_pwa_audit_statusget_pwa_audit_resultslist_pwa_auditscompare_pwa_auditsget_pwa_audit_entitlement
Example requests:
Run generic PWA checks on https://example.com. Do not start a hosted audit.Run a standard PWA Today audit on https://staging.example.com. Use this repository and commit as source metadata. Show me the result when it finishes.Review audit <auditId>, map every failed or warning check to this codebase, and propose a minimal fix plan. Do not edit files yet.After the deployment is live, run the audit again and compare the quality gate and failed checks.List the 10 most recent audits for example.com and identify the latest completed run.Compare these two audit IDs: [baseline audit id] and [candidate audit id]. Tell me which checks resolved, which regressed, and whether the comparison used the same ruleset.Check my PWA Today entitlement before starting an audit. Do not start one yet.check_pwa accepts public HTTPS targets only. It does not create an audit, require an account, or consume allowance. start_pwa_audit creates a hosted audit and still requires a verified application hostname. The server creates an idempotency key for each new audit and reports it with the audit ID. If creation has an unknown outcome, retry with that same key. The MCP package does not accept audited-application authentication or deploy applications.
create_pwa_audit_configuration saves a reusable configuration without running an audit. Agents should retrieve entitlement first, guide the user through the plan-compatible checks and required options, and confirm the final configuration before saving. It does not accept site authentication or other secrets.
Listing audits, comparing audits, and retrieving entitlement do not consume allowance. Starting an audit can reserve or consume allowance according to existing API behavior. Comparison reports observed result differences; it does not prove that a particular code or deployment change caused them.
This server cannot be deployed
Maintenance
Related MCP Connectors
Operate smplkit from your agent: feature flags, config, logging, audit, and scheduled jobs.
Remote MCP for Android CLI agent build gate, structured receipts, audit logs, and reviewer-ready evi
Control plane for autonomous software labor. Agents claim objectives over MCP with audit trail.
- mcpOAuthio.artifacta
Artifact store for AI agents. Hosted OAuth at mcp.artifacta.io/mcp; local stdio via npm/PyPI.
Related MCP Servers
- AlicenseAqualityCmaintenanceExposes axe-core accessibility auditing as MCP tools so any MCP-compatible agent can audit HTML for structural and semantic issues.3MIT
- AlicenseNot gradedqualityBmaintenanceDelegates bounded coding tasks from MCP clients to the Pi Coding Agent over stdio. Supports review, verification, implementation, and batch operations with long-running task polling.MIT

yotta-verify-mcpofficial
AlicenseNot gradedqualityBmaintenanceProvides a deterministic pre-install security scan for Agent skills, plugins, and MCP servers via stdio, returning verdicts, audited badges, CI gates, and reports without executing code or making network calls.707 npmMIT- AlicenseNot gradedqualityAmaintenanceExposes ReadyAgents' local workflow engine and built-in tools over MCP stdio, enabling MCP clients to define, run, inspect, and approve YAML/JSON agent workflows.919 PyPIApache 2.0