OAuth-backed app connections
Connect the apps an agent is authorized to use through OAuth-backed connectors, without placing service credentials in the agent.
CoreSpeed
CoreSpeed is an MCP-based access layer that gives AI agents authorized app connections, portable memory, built-in media and research tools, and spending controls. It is designed for individuals and teams using cloud, local, or custom agents across workflows such as sales, marketing, support, engineering, and operations.
Agent access, memory, and controls in one MCP endpoint
CoreSpeed is an MCP-based access layer for AI agents. It gives agents such as Claude Code, Codex, Cursor, cloud agents, local agents, and custom agents access to authorized applications, account-level memory, and built-in tools for media, web search, scraping, and social research.
The service is intended to make an agent setup portable across tools and execution environments. Users authorize connectors through OAuth, configure action and spending controls, and review activity through a shared account rather than managing separate credentials and memories for each agent.
Core capabilities
Shared MCP access layer
Expose connected apps and built-in tools through one authenticated Model Context Protocol endpoint for supported and custom agents.
Built-in research and media tools
Provide media generation, image, video, and audio analysis, web search and scraping, and social research tools from the same agent access layer.
Portable agent memory
Store durable memory on the CoreSpeed account so it can be available when users move between agents rather than remaining tied to one machine.
Action and spend controls
Apply plain-English approval policies, handling rules for unspecified actions, and spend caps before paid calls are made.
Activity logging
Record activity in a ledger showing what ran, what it cost, and whether it worked, with plan-dependent history retention.
Practical use cases
Sales and revenue workflows
Research prospects, draft outreach, and log customer relationship activity through connected sales applications without copying information between tools.
Marketing and growth operations
Build campaigns, schedule social content, and retrieve funnel information through authorized marketing applications and social research tools.
Customer support triage
Triage support tickets, prepare replies using available context, and identify potential churn signals as part of a support workflow.
Engineering and DevOps coordination
Review engineering backlogs and pull requests, then manage sprint closeout from an agent conversation connected to development tools.
Finance and operations
Categorize expenses, reconcile payouts, and support month-end close processes while applying approval rules and spend limits to agent actions.
Pros and Cons
Pros
- One authenticated MCP endpoint can be used across multiple agent products and custom agents.
- OAuth 2.0 connectors provide scoped access without giving CoreSpeed service passwords.
- Durable memory and app connections can follow a user when switching agents.
- Approval policies, spend caps, and an activity ledger provide controls for unattended or team-based agent workflows.
- Supports both individual use and organization plans, including custom enterprise options.
Cons
- Usage is metered in CoreSpeed credits, and some connectors and built-in tools have per-action or per-request rates.
- The amount of memory, activity history, included credits, and team capacity depends on the selected plan.
- Connected services still require authorization and may impose their own access constraints.
FAQ
How do I set up CoreSpeed?
Give the agent the setup line `set up https://corespeed.io/SKILL.md` and have it read the file raw, for example with curl. The setup process adds the CoreSpeed MCP server, signs the user in or creates an account, and keeps the skill available for later use. Manual setup instructions are available in the CoreSpeed documentation.
Does CoreSpeed receive my service passwords?
No. CoreSpeed uses OAuth 2.0 authorization through each provider's own authorization screen. It receives scoped tokens rather than service passwords or credentials.
Which agents can use CoreSpeed?
CoreSpeed is designed to work with Claude Code, Codex, OpenClaw, Hermes, Cursor, cloud or local agents, and custom agents. Its MCP endpoint acts as the shared access layer.
What plans are available?
CoreSpeed offers a free plan with 3,000 credits valid for 90 days, one member, 200 MB of memory, and 30 days of activity history. Pro costs $20 per month and includes 10,000 credits per month, up to 25 members, 2 GB of memory, and one year of history. Enterprise pricing is custom.
How can I control agent actions and spending?
Users can write approval policies in plain English, choose whether unspecified actions run, request approval, or are refused, and set spend caps. Actions are recorded in an activity ledger, and approvals can be requested in the dashboard, by email, or in Slack.
Quick Facts
- Category
- AI agent infrastructure
- Access method
- Authenticated Model Context Protocol (MCP) endpoint
- Supported environments
- Cloud, local, and custom agents
- Built-in tools
- Media generation, media analysis, web search, scraping, and social research
- Free plan
- $0; 3,000 credits per 90 days, one member, 200 MB memory
- Pro plan
- $20/month; 10,000 credits per month, up to 25 members, 2 GB memory
Alternatives à CoreSpeed
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
AakarDev AI
AakarDev AI helps teams manage AI provider access, project-level setups, logs, and analytics from one dashboard. It supports BYOK workflows and lists providers including OpenAI, Google Gemini, Anthropic, Groq, Mistral AI, and Perplexity AI.
Trigger.dev chat agent
Trigger.dev chat agent is a durable AI chat backend for developers building stateful conversations that can survive refreshes, crashes, and long-running turns. It connects with the AI SDK `useChat` flow and runs on managed infrastructure with no timeout on a turn.
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
Codex Plugins
Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI. They help extend Codex with connected-service tasks, reusable instructions, and shared team workflows.