Agent Context Needs an Import Statement
Engineering Blog · September 7, 2026
Apply modules, dependencies, and imports to agent context. Mike Morton explains the SemDAG architecture and how Cortex makes shared knowledge available across agents.
Cortex lets you and your team share context across your AI agents, giving each agent the right knowledge before it acts.
Connect through Cortex plugins, MCP or the Socra CLI.
Your systems carry decisions about what matters and why. When that knowledge is missing, agents have to make assumptions. Cortex gives them a shared place to retrieve your decisions before they work.
Cortex gives agents your specific knowledge and its reasons before the work begins.
Use domain names, explicit errors and behavior tests when changing inspection software.
Name observations, visits and reports consistently. Distinguish access denial from an unavailable connection. Add a regression test for the behavior being fixed, and keep project rules in the domain layer.
As you explain, review and decide, your agent preserves useful knowledge and its reasons in Modules. You keep doing the work.
The agent maintains the dependencies between related Modules. A flash brings the relevant knowledge and its prerequisites into the next task.
Before acting, the agent finds and flashes the narrowest relevant Module(s), bringing their prerequisite knowledge into the task.
The agent reads your reasons, verifies the current code and runtime, then works against your standard. You review whether the result follows your system.
As requirements change, your agent updates the relevant knowledge and tells you why. Choose automatic updates or approval before changes are saved.
An Issue says what must change. Its Module says what must remain true.
your inbox
released work
Recompute tax when the address changes
#418
checkoutReject expired promo codes in the cart
#392
pricingShow an ETA for partial shipments
#377
shippingEvery Issue names its Module. The agent or teammate who picks it up can flash that Module before work starts.
Issues declare dependencies on other Issues. The inbox releases an Issue after all its dependencies close.
The inbox shows released, unclosed work assigned to you.
This recorded CLI call retrieves a visit-report Module and its writing prerequisite. The example knowledge is fictional; the retrieval is real.
socra cortex flash example-fieldnote-visit-reportThe writing prerequisite arrives first:
Fieldnote visit reports use sentence-case titles and plain language. Explain unfamiliar terms on first use, so a colleague who did not attend can understand the report.
Then the requested visit-report Module:
A visit report names the visit’s date and location, separates direct observations from interpretations and ends with the next action and its owner. Keep that distinction because a later reader needs to know what was observed and what still needs checking.
Models can apply new knowledge without being retrained. Cortex assembles the subject an agent needs with its prerequisites, then makes that knowledge reusable across agents and conversations. Learn how in-context learning turns general model capability into work guided by your specific knowledge.
Your Modules live in the cloud as one graph: stored in your Account, addressed by name and owned by you. Any authorized team member or agent can flash the same source from a connected tool.
A Module is referenced by name from repositories and tools connected through Cortex.
The Account boundary isolates its people, installed agents and Modules from every other Account.
Every Module has an owner. Modules and Issues record who changed what, person or agent.
Start alone and add members when you need them.
An Account is the tenant, billing and isolation boundary for its people, installed agents and Modules.
People and agents in an Account work from the same Modules.
Account
acme-platform

you
Owner
Dana
Member
Priya
MemberChatGPT
Installed appAccount boundary · isolated tenant
You don’t have to write Modules before you begin. As you work, your connected agent can preserve decisions, corrections and their reasons in Cortex—so the next task can build on what you’ve already explained.
Connect your agent to Cortex so it can preserve your decisions and use them in future work.