Launching today
Graft AI
Turn company operations into a living map for agents
36 followers
Turn company operations into a living map for agents
36 followers
Most agent tools assume clean APIs. Graft starts where companies actually work: legacy apps, internal tools, and workflows trapped behind screens. It learns how the work gets done, turns it into a living operational map, and gives agents stable tools with permissions, approvals, audit trails, and verification built in. When the underlying UI changes, Graft detects the drift and repairs the workflow without breaking the agent interface.
Graft AI
if the underlying business logic changes and not just the screen, does the agent, facing tool silently start doing the wrong thing instead of breaking?
Graft AI
@derek_julian nope, Graft treats a business-rule change as semantic drift, not a UI repair. Every tool is bound to explicit policies, expected effects, and independent verification; if the source system no longer behaves as certified, the tool fails closed and is quarantined until a new version is reviewed and re-certified. I previously worked on enterprise infra where data was messy and agents needed it in real time, this was the exact problem I faced and I've spent my time solving this
nothing in the pitch about workflows that need a human judgement call partway through, is that out of scope for now?
Graft AI
@leah_josephine No. Human judgment is a first-class step in Graft workflows. Agents run autonomously until they hit a policy boundary or a genuinely ambiguous exception, then route a concise, evidence-backed decision to the right person and pause only that step. Once decided, the workflow resumes from the same state, so humans handle exceptions rather than babysitting every run.
building stable tools instead of chasing every API makes sense, but that shifts a lot of ongoing maintenance onto graft's side to keep those maps accurate as companies change.
Graft AI
@dustin_warren Exactly, which is why maintenance is part of the product: Graft continuously detects UI and business-logic drift, fails closed when a workflow no longer matches its certified behavior, and re-verifies every update before agents can use it again.
the framing that the model was never the hard part is honestly accurate for most companies, the real question is whether the "living map" stays correct as workflows quietly drift over months, not just when the UI visibly changes.
Graft AI
@evan_cooper1 yes, Graft continuously compares real decisions, exceptions, and outcomes against the certified workgraph, so quiet semantic drift is detected and reviewed before it becomes production behavior. We are also building a workflow machine that can replay any historical case under today’s rules and show exactly where the company’s operating logic changed.
One thing that would help us adopt this faster is a visual timeline view inside the operational map, so we can scrub through past workflow executions and see exactly where drift or failures happened. Right now troubleshooting at scale feels like guesswork.
Graft AI
@alpergrbzobpev Yes, that is exactly where Graft is headed: every run becomes a replayable visual timeline, so teams can scrub through decisions, handoffs, failures, and drift instead of debugging from scattered logs. Your company owns the resulting operational knowledge base, and every employee correction makes it smarter, much like a team continuously improves a shared codebase for agents. If you're interested and want to talk, The waitlist is live at graft.axcelner.com.
feels adjacent to Browser Use in spirit, just aimed at ERPs and desktop apps instead of the open web.
Graft AI
@ian_maxwell2 No actually, this is really different. We're not teaching agents how to interact, we're building an operational knowledge map of the company, how operations really work and today it is stuck in people's head as domain knowledge, we saw a shift where now anybody can contribute to a company's codebase but imagine now employees knowledge and how work is done in the company is operatable by agents consistently, this still breaks at scale today and we're solving that.
Browser Use helps an agent operate a UI, while Graft turns recurring enterprise work into versioned, policy-bound, independently verified tools that agents can call reliably without rediscovering the interface every time.