Early access · In production

Open source agents for your Company Brain

Minerva Stands is the auditable runtime for governed operating instructions. It connects Company Brain context to agent work with active guardrails, path-level access control and embedded audit logging. Official Docker images for deployment — coming soon.

Minerva Workbench
Minerva Workbench: file explorer, GUARDRAIL.md editor and agent chat
Agent UI preview — inspect the workspace, chat with the agent, and query the audit log before operational work runs

Internal architecture

Diagram: containerized layout with interaction layer, heartbeat with guardrails, file-based context, and SQLite audit (official Docker images coming soon)
Target architecture: chat apps and the UI, heartbeat governed by GUIDERAIL.md, Company Brain context in files (MEMORY, USER, SOUL, skills), and a SQLite audit database. Official Docker images — coming soon.

The execution layer for governed company knowledge

A Company Brain gives agents the operating instructions. Stands is how those instructions can run with guardrails, workspace limits, review and audit. We start from nano-bot, the autonomous agent from the University of Hong Kong, and harden it with four Minerva layers so it can operate in real environments without surprises. Everything is open source. Official Docker images for deployment are coming soon.

Built on nano-bot (HKU) 100% open source · Permissive license
  • 01

    Active guardrails

    Rules, workflow context and Company Brain constraints injected into every prompt.

  • 02

    Path access control

    Allow/deny list of editable paths.

  • 03

    Full auditability

    Every interaction is normalized and stored in a SQLite database embedded in the agent itself.

  • 04

    Operator UI

    Built-in UI to inspect the workspace, chat with the agent, validate intermediate actions, and query the SQLite audit log — all in one place.

Four layers between the Company Brain and automated work

Agents should not improvise from raw documents. Stands keeps model actions inside reviewed instructions, allowed paths and an audit trail.

01

Active guardrails

Rules, workflow context and Company Brain constraints injected into every prompt. The agent can't ignore them because they travel with each model call: tone, scope, forbidden actions and goals, always present.

02

Path access control

Allow/deny list of editable paths. You define what the agent can read, write or ignore, so Company Brain instructions cannot become broad filesystem access.

03

Full auditability

Every interaction is normalized and stored in a SQLite database embedded in the agent itself. Full traceability: Company Brain context, prompts, responses, tools used and file changes.

04

Operator UI

Built-in UI to inspect the workspace, chat with the agent, validate intermediate actions, and query the SQLite audit log — all in one place.

Where Company Brain instructions turn into work

Three production deployments with Minerva customers—not slideware. Each agent owns a workflow and acts from governed context: calendar and inbox, social content, or customer memory.

Live

Calendar & email manager

A digital secretary with access to the calendar and inbox. Prioritizes and categorizes emails based on whether they need action, oversight, are blocking something, or are purely informational.

  • Instant alerts for high-priority actions
  • Bulk-edit queue for the rest, sorted by category
  • Drafts ready-to-send responses for your review
Live

Autonomous marketing manager

Once a day it scrapes news relevant to the company, extracts topics with cited sources, and after user approval writes, reviews and schedules content across social networks.

  • Daily topics with verifiable sources
  • Automated post drafting and review
  • Generated infographics and multi-channel scheduling
Live

Customer Company Brain

Active memory with a custom protocol and Minerva's internal tooling. Categorizes and keeps every customer interaction, demo and commitment up to date, wired into Google Calendar.

  • Active per-customer memory with accumulated context
  • Creates events and blocks on Google Calendar
  • Generates internal briefings so the whole team stays aligned

Open code, deployed around your knowledge

We believe autonomous agents should only run when the knowledge, instructions and actions can be audited. Minerva Stands is open source end-to-end. Official Docker images are coming soon so you can inspect, modify, and choose where it runs.

View on GitHub
  • 100% open source — audit every line, no black boxes
  • Official Docker images — coming soon (deploy on your cloud or ours when released)
  • No vendor lock-in: keep it and operate it yourself if you want

Ready to connect agents to your Company Brain?

Tell us the process you want to automate. In 30 minutes we'll tell you whether it needs better knowledge, stronger governance, or an auditable Stands workflow.