What ContextWorks adds to Salesforce.
The platform provides the means to connect your AI to the data. What that data means is still yours to define.
Start with what's already there.
Context and semantic are words that are used a lot in the market right now, and several Salesforce features are described with these words. These are the pieces the platform has, and what each one actually provides.
- Hosted MCP Servers
- Secure, governed, per-user access to MCP tools for AI tools and agents outside Salesforce. Every call runs as the authenticated user. This is the access problem, solved securely.
- Data 360
- The data platform. It unifies and models data from across your systems, and powers unstructured grounding, indexing and observability for Agentforce agents. Grounding any agent, inside or outside Salesforce, in CRM records does not require it.
- Tableau Semantics
- A real semantic layer, and a good one. It defines metrics once — what revenue or churn means — consistently for analytics and for Agentforce agents. It is bound to Data 360 and oriented to metrics and aggregates.
- Agentforce Observability
- Session tracing over an Agentforce agent's reasoning, so you can see how it arrived at an answer. Strong on the reasoning; it is not a view of the context payload assembled for a given record and user, and it does not cover agents outside Salesforce.
- Field Descriptions
- Salesforce's own MCP servers return each field's description to the agent calling them. That is the description the field was set up with in Setup, written for admins. There is no way to curate it or define it for an LLM, and it reads the same in every use.
- Context Service
- Worth naming because the word collides. This is an Industries developer feature — an in-memory data context that saves repeated database trips for Flow and the Business Rules Engine. It is a performance mechanism, unrelated to AI.
We use the platform rather than rebuild it.
ContextWorks is a Salesforce-native application. Permissions, sharing and the trust layer are not features we reimplemented — they are the platform's, and we run inside them.
| Field-level security and sharing | Every query runs in user mode, so permissions are enforced on the query itself rather than filtered afterward. |
| Hosted MCP Servers | ContextWorks can be deployed as its own MCP server or embedded in other ones, using the platform's OAuth and registry. |
| Einstein Trust Layer | Agentforce delivery flows through it unchanged. |
| Prompt Builder, Flow and Apex | Grounding for templates, invocable actions for automation, and a public API for your own code. |
| Field metadata and page layouts | Entities initialize from your existing setup rather than a blank canvas. |
| Data 360 | Calculated Insights and DMOs surfaced as enrichments on a record. |
Full context, built by the people who know the data.
ContextWorks provides the layer between your data and your AI, where a field gets a meaning, a definition gets an owner, and every AI tool gets the full context it needs.
Entities
The records, curated
The fields that matter for each record, at three levels of detail, each with an instruction. Rollups, related records and computed signals come with them.
Briefs
The knowledge that isn't in any field
The business knowledge outside the data: how you segment, what a stage means, your voice and your policies. Written once by the team that owns it.
Skills
How the work gets done
The steps your teams follow, provided as procedures any agent can use. They allow agents and AI tools to collaborate and handle tasks effectively.
Ownership
Maintained by the business
Each team maintains its own domain, with delegated administration, versions and one-action rollback. No sandbox or deployment required.
One definition, every surface
Built once, read everywhere
MCP, Agentforce, Prompt Builder, Flow and Apex all read the same definitions, so the answer is the same in every tool.
