Product updates

Ticketing integrations: answers ready before IT support even starts on the ticket 

A single M365 access ticket means jumping between the ticketing system, admin centers, and audit logs. Support teams lose time searching across disconnected systems and tribal knowledge, and new staff take a long time to get productive. 

Syskit Point integrates with the ticketing system your team already uses. When an M365-related ticket arrives, Point automatically attaches an internal note with the answer — for example, "John lost access because Ana removed it on 25 May" — so the agent starts with the answer already in hand. Native connectors are available for Jira, ServiceNow, Zendesk, Freshdesk (Freshservice), and Halo, and you can integrate almost any ticketing system that supports webhooks and APIs. 

Why it matters

• Faster resolution - the answer is attached before IT support opens the ticket
• New staff are productive without the tribal knowledge
• Better CSAT

MCP server: put Point's intelligence into the AI tools you already use 

As teams route more work through AI tools and agents, governance intelligence locked inside a single product UI becomes a bottleneck.  

Syskit Point's MCP server exposes its data and capabilities as a standard interface — the Model Context Protocol — that your own AI tools can connect to and act on. In practice, everything Point Assistant can do becomes available outside Point, not just inside it. An admin can ask Claude or Copilot and get the answer without opening Point, and custom agents can pull the same intelligence into their workflows. 

It unlocks lookups across users, groups, teams, and sites, running and interpreting Point reports, querying audit history, analyzing storage and version-cleanup opportunities, and inventorying Power Platform resources and their ownership risks. 

Why it matters 

• Point's intelligence reaches the AI tools your team already uses, not just Point's UI

• Admins get answers by asking their assistant — no need to open Point 

Point Assistant: ask governance questions, get answers 

Answering everyday questions — who has access to what, why a user is risky, what an account has been doing — usually means knowing which reports to run and where each setting lives. 

Point Assistant turns governance work from "go dig through dashboards" into a conversation. Ask in plain language and get answers straight from Point's own data. It can already find and review users, groups, and SharePoint sites, check access and memberships, analyze storage, investigate audit activity, surface tenant insights, run reports, review Power Platform items, and answer product guidance questions.

Why it matters 

• Answers in seconds, without knowing which report to run

• Plain-language findings, not log dumps

• Faster, more consistent access reviews 

Workspace summaries: a plain-language health check in one click 

Judging whether a single workspace is healthy means digging through dashboards and cross-referencing sharing links, permissions, sensitivity labels, inactive access, and storage. In an environment with thousands of workspaces, admins sometimes have no time for that, so risks sit unnoticed until they become incidents. 

Now, a single click on "Summarize" produces a plain-language overview: the top 3-5 issues that need attention, benchmarked against your organizational averages, with the impact of acting quantified (for example, more than 100 GB reclaimable by clearing old file versions). Each item links directly to the report where you fix it. 

Why it matters

• See what really needs attention, not everything at once
• Evidence behind each finding - every finding links straight to its fix
• Assessment time drops from hours to seconds

Sensitivity label recommendations: close the classification gap 

Unlabeled or wrongly labeled workspaces are open exposure windows - policies and protections simply don't apply to them. Manually classifying every new and legacy workspace does not scale. 

Machine learning analyzes the workspaces you have already labeled, recommends labels for the rest, and flags existing labels that look wrong. Recommendations improve as coverage grows, and you stay in control: every recommendation is confirmed, never auto-applied. 

Why it matters 

• New and unmanaged workspaces get classified sooner
• Existing mislabels are surfaced, not just gaps
• On by default, ML only, no opt-in needed

*A note on data privacy and security: AI features are off by default, LLM processing stays in your Azure boundary, and your data is never used to train any models.