Retrieve project context.
Read relevant Brain files and retrieve published organization skills before starting work.
MCP & plugin
Your agent environment. Your team’s context.
MCP gives your agent access to Brain files and published skills. The Venom plugin adds guidance to retrieve project context and record useful findings during the work.
The MCP connection
A direct connection to your workspace.
The Model Context Protocol (MCP) exposes the operations an agent needs to work with your Venom filesystem, within the permissions of the user it represents.
Add a finding.
Create a task note, record an experiment result, or document an implementation constraint.
Update the working record.
Update an existing file as the project changes, so future sessions use the revised context.
The session routine
Project context, built into the session.
On supported hosts, the plugin reminds the agent to consult Venom at session start and when work resumes. Its guidance also asks the agent to save useful findings before handing the task back.
01 / Read
Read the experiment notebook.
The agent finds the attention-window hypothesis, code map, and fixed compute budget in Brain.
brains/alex/tasks/attention-window/todo.md02 / Work
Apply the experiment method.
It retrieves the published skill: establish a baseline, hold evaluation fixed, and compare the candidate.
Published skill / run-an-experiment03 / Write back
Save the result with the task.
It records the run’s code revision, validation score, and outcome. The next session can inspect the evidence.
brains/alex/tasks/attention-window/experiments/results.mdAccess follows your role
Your agent works with your permissions.
MCP access follows the authenticated user’s permissions. Your agent can read or change files only where your permissions allow it. Owners and admins publish reusable skills for the team. Each agent works with the knowledge its user is allowed to access.