Snapshot
Continue to About ↓- What it does
- Scalekit AgentKit is a managed MCP gateway that lets your AI agents act inside third-party SaaS apps (Gmail, Slack, Salesforce, GitHub, Notion, Jira and others) on behalf of each end user.
More detail
- Each user connects their own accounts once
- Scalekit stores and refreshes the OAuth tokens and attaches them to every tool call, so your code never handles credentials.
- It also adds scoped tool access per agent, event triggers, and per-call audit logs.
- Best for
- Product and engineering teams building or hosting their own AI agents that need to act inside customers' SaaS accounts, and who want to keep their existing agent framework while outsourcing the OAuth and connector layer.
More detail
- Also fits enterprises with security review requirements such as BYOK, VPC or self-hosting, and US/EU data residency.
- Pricing
- Free tier at $0/month, but the published plans table lists 5,000 tool calls per month as a hard stop, which conflicts with the plan card's 10,000 tool calls and 10,000 trigger events.
More detail
- Growth is $99/month with 100,000 tool calls and 50,000 trigger events included, then $0.50 per 1,000 extra tool call
- Main limitation
- Teams that need only a few stable integrations, where a handful of direct API calls may be cheaper and simpler than adopting a gateway.
More detail
- Not for buyers looking for a financial or trading product; AgentKit is a software platform only.
- Record
- Living and monitored
- Reviewed
- Evidence
- Public web research
- Machine access
- Structured profile ↗
About
- Scalekit AgentKit is a managed MCP gateway that lets your AI agents act inside third-party SaaS apps (Gmail, Slack, Salesforce, GitHub, Notion, Jira and others) on behalf of each end user.
- Each user connects their own accounts once
- Scalekit stores and refreshes the OAuth tokens and attaches them to every tool call, so your code never handles credentials.
- It also adds scoped tool access per agent, event triggers, and per-call audit logs.
What it helps with
- Building and maintaining per-connector OAuth flows and token storage in your own codebase
- Risk of one user's agent reaching another user's connected accounts
- Reconstructing which agent, user and tool caused a failure from scattered application logs
- Owning polling, cron jobs or webhook infrastructure to start agent runs on events
- Loading too many tool definitions into an LLM context, causing wrong tool calls
- Meeting enterprise security review requirements for credentials, residency and deployment
Capabilities
A practical view of what is available and where limitations apply.
| Offering | Available | Detail |
|---|---|---|
| Connector catalog | Yes | Official site claims 600+ connectors and 30,000+ tools; product updates say the catalog crossed 10,000 tools, so verify the current count. |
| Managed per-user OAuth tokens | Yes | Tokens encrypted per tenant, refreshed before expiry, and attached to each call; your code and prompts do not hold them. |
| Scoped tool access per agent | Yes | The gateway enforces which tools each agent can see, rather than relying on prompt instructions. |
| Virtual MCP Servers | Yes | Compose a role-specific tool set from multiple apps behind one endpoint; vendor claims about 80% less tool overhead. |
| Event triggers | Yes | Start agent runs on events such as a new email or a deal stage change, using the user's connected accounts. |
| Custom tools and bring-your-own MCP server | Yes | Add internal or partner APIs and extend connectors; supports OAuth 2.0, API keys, bearer and basic auth. |
| Per-call audit logs and SIEM streaming | Yes | Every call attributed to agent, user, tool and result; streamable to Datadog, Splunk or any SIEM. Full streaming is an Enterprise feature. |
| BYOK, VPC and self-hosted deployment | Yes | Enterprise options include AWS/GCP KMS keys, VPC or self-hosted, air-gapped support and IP allowlisting. |
| SOC 2 Type II and HIPAA BAA | Yes | SOC 2 Type II reported; HIPAA BAA and custom DPA available on Enterprise only. |
| Independent third-party reviews | Unknown | Only a LinkedIn rating of 4.9/5 from 23 reviews was found; no independent review sites or press coverage located. |
Bot trust
How confidently AI systems can verify this profile from public evidence.
—of 100
ChatGPT—/100
Claude—/100
This profile has not completed a Bot Score assessment yet.
Evidence
Where this record came from, when it was reviewed, and how confidently it can be used.
- Verification
- Scribble researched these facts from the linked public sources.
- The owner has not manually confirmed every field.
- Last reviewed
- Primary source
- scalekit.com
Technical profile details
Record ID
d0ced83b-5861-4cff-a629-3f2e2a5db0b7Canonical URLhttps://profiles.scribble.network/listing/scalekit-agentkit-d0ced83b