Autonomous agents of the class now coming for law firms, pointed at your platform first — as every role and at more than one firm, with proof attached to every finding and the evidence your SOC 2 and ISO 27001 audits ask for.
#securityandsecurity@acme.legalnotified
The boundary your whole product rests on, and the one an object id walks straight through. Agents hold accounts at separate firms at once and work every pair between them: matters, documents, timelines, billing.
A conflict screen is an access rule, and access rules get enforced where they are easiest to write. Agents sign in on both sides of a wall and check whether the screened matter is genuinely unreachable or merely hidden from the menu.
Legal work runs on links sent to opposing counsel, clients and experts. Agents test what a link still opens after the matter closes, after the recipient is removed, and whether it grants more than the one document it was meant to.
Bulk download, e-discovery production, the audit-ready PDF. Export paths are written late and reviewed less, and they routinely return more than the user could reach through the interface.
Every domain you own, the providers underneath, and the hosts nobody has thought about since they were created. Discovery finds them from the outside, so a forgotten staging copy of the document store stays on the map whether or not anyone remembers deploying it.
Nothing is recorded without the requests and identities that proved it.
Connect it over MCP and point it at the finding. The exploit is the spec.
It reproduces, or it does not. That is the answer your customers' counsel wants.
Point agents at your platform as every role you have, at more than one firm, and see what crosses the boundary your product is sold on. Every finding arrives with the request that proved it.
Common questions from teams building legal platforms.