Read-only by default
A connector starts with read access. Write scopes are added deliberately, per object and per field, and are visible in the console.
Integrations
Read almost everything. Write almost nothing, and only with permission.
Mayeshal reads broadly and writes narrowly, on purpose. Every connector enforces the calling user's own permissions in the source system, so an agent can never write where the person could not, and every write back into your case system is staged, whitelisted and approved by a human before it lands.
The contract
Four rules that apply to every connector.
Your permissions, not ours
Access is granted through an authorisation in your own system, held per tenant and encrypted. You control the permission set and the object and field scope, and you can revoke it from your own administration screen at any time without contacting us.
Field-level access controls
Beyond what your system enforces, connectors carry their own field-level allow lists, and metadata paths that should never be traversed are blocked outright.
Sandbox before production
Connectors are environment-aware. We connect to your sandbox first, validate against your real configuration, and only then authorise production.
The library
What we connect to.
Three lists, and we will always tell you which one you are looking at. If a system is in the first, we can show it to you working. If it is in the second, we can show you the code. If it is in the third, we can show you the engagement.
Running in production today
Salesforce platform applications, including Litify
Read across every object on the record: matters, intakes, parties, damages, insurance, liens, negotiations, tasks and calendar events, with staged writeback that a person approves. This is the connector we can show you working against your own data on the call.
The document system behind that platform
Every document read page by page, indexed, summarised and cited, then held in your tenant's own storage after first read so it stays openable when the source system is not.
Meeting transcription
Fireflies and equivalent services, so meetings become searchable alongside documents rather than living in a separate tool.
Built, and authorised during implementation
These connectors are written and tested against the platforms they target. Authorising one in your own tenant is an implementation week task, not a build.
Microsoft 365 mail, through Microsoft Graph
SharePoint document libraries
Filevine
QuickBooks
NetSuite
A document execution platform
Systems we work in every week, built as a connector on request
These are not shipped connectors and we will not pretend otherwise. They are systems Micronetbd implements and administers for clients, which is why scoping a connector to one of them is a conversation rather than a research project.
DocuSign, including agreement lifecycle management
PandaDoc
SAP, including S/4HANA
Specification and product data platforms
Loan origination and core banking
Contact centre platforms
Business messaging platforms
Google Workspace
Slack
Anything with a documented API
If a system has a documented REST or GraphQL interface, it can be built as a custom connector. Scope and price are quoted per connector.
After the connector
Read once, held safely, openable forever.
The first time we read a document we hold a copy in your tenant's own object storage, private and versioned. Three things follow from that. Opening it later takes about three hundredths of a second instead of one to two seconds. It stays openable when the source system is throttled or down. And a freshness job checks the source nightly, so if the original changes, our copy is marked stale and re-read rather than quietly diverging.
Every stored file is checked against its own magic bytes on both write and read, because document systems will occasionally hand back a PDF wearing a Word document's filename, and a cache that accepts that is a cache that lies to you later.
Not in the list
If it has a documented API, it can be connected.
Line-of-business systems, custodial feeds, portfolio accounting, records retrieval vendors, medical chronology providers. Bring the API documentation to the call and we will scope it and quote it. What we will not do is quote a connector before we have read the documentation, because that is how a two-week build becomes a two-month one.
Get started
See it read one of your own matters.
Thirty minutes, on a screen share, on a matter you pick. We connect nothing, install nothing and ask for no credentials to have that conversation. If it is not obviously useful in the first ten minutes, we will tell you so and you get your afternoon back.