AI agents run your projects. You make the decisions.
Nova and your specialist agents plan sprints, triage tasks, and keep docs current — every action logged, reversible, and yours to approve.
EU-hosted · Self-host ready · SHA-256 audit trail · Free forever — no card
MKT-128 · agent run · 3.1s
Plan the Q3 product sprint
sha256:9f2ac1… · reversible
- 80+ typed agent tools
- ·100% EU-hosted
- ·7 entity types, one-click rollback
Every action written to a tamper-evident audit trail.
Every action audited. Every decision yours.
Nova drafts the plan
A specialist agent picks up the work, breaks it down, and proposes every change.
Approve, edit, or roll back
Nothing high-risk executes without you. Configure the rest through per-agent approval rules.
Signed to the audit trail
Every applied change is written to a tamper-evident, SHA-256 audit chain — reversible at any time.
Your team. Your agents.
One orchestrator, Nova, coordinates specialists — a sprint planner, docs writer, QA engineer — that you configure and control.
Custom system prompts · scoped tools · skill files · memory that learns from every run.
Reads incoming work, picks the best specialist, delegates, and reports back. Built into every workspace. Built into every workspace as your central coordinator.
Turns a brief into a multi-channel campaign plan — timeline, deliverables, and owners.
Reads tickets, checks the account, drafts replies, and writes help-center articles.
Enriches inbound leads, finds the stakeholders, and assembles an account brief.
Drafts personalized outreach and logs every touch to your CRM.
Builds new-hire checklists, schedules week one, and drafts the welcome docs.
Turns a product brief into epics, stories, estimates, and a dependency graph.
Brief an agent. Watch it work.
Ask in plain language; Nova routes it to the right specialist and streams its reasoning as it goes.
- –Nova orchestrates the right specialists
- –Live reasoning and tool calls
- –Every run logged and reviewable
Work your agents actually manage.
Agents create, de-duplicate, estimate, and advance tasks across board, list, and timeline — linked to the commits that resolve them.
- –Board, list, and timeline views
- –Auto-estimated and de-duplicated
- –Linked to the commits that resolve them
Docs that maintain themselves.
Docs agents write and update articles, build courses, and keep everything linked to your work. Publish internally or as a customer-facing site.
- –Rich editor and knowledge graph
- –Sites with your own branding
- –Linked to tasks and features
Channels tied to the work.
Standup summaries, blocker detection, and re-plan suggestions, threaded to the tasks they're about.
- –Agent-summarized standups
- –Threads linked to tasks
- –Channels, DMs, and mentions
QA your agents run.
Agents generate test cases from requirements, run suites, and trace results back to the spec.
- –AI-generated test cases
- –Suites and tracked runs
- –Full requirement traceability
You set how much autonomy each agent gets.
An approval broker sits between every agent and every action. High-risk actions always ask; you decide which of the rest need sign-off — per agent, per tool.
Copilot
You approve everything. Each proposed action waits for your sign-off before it runs.
Autopilot
Low-risk tools run on their own; you review the results. High-risk actions still stop for approval.
Custom
Choose which actions need your approval, per agent and per tool. Set a confidence threshold and anything below it is held for you.
High-risk tools always require confirmation. Confidence scores are recorded and shown — they inform your thresholds, they never auto-execute.
Trust through transparency, not promises.
Every agent action is logged, auditable, and reversible. You're never wondering what happened or why.
Tamper-evident audit trail
Every action is written to a SHA-256 hash chain with its full reasoning. Break any link and verification fails — accountability you can prove.
One-click rollback
Every change captures a before-and-after snapshot. Don't like what an agent did? Roll it back and the state is restored.
Confidence scoring
Each action records how confident the agent was and why. Scores are shown, not auto-executed — anything below your threshold is held for approval.
Run It Your Way
Your data, your infrastructure, your rules. Or let us handle it.
Self-Hosted
Docker Compose, one command. Bring your own LLM. Air-gap ready.
Enterprise
Coming nextHelm chart, SSO, pen-tested, SLAs. For regulated industries.
Built for the teams that run the work.
Same engine, different playbooks. Point your agents at the work your team actually does — you stay in the approval seat.
Engineering Teams
Engineering TeamsAgents handle sprint planning, task management, documentation, and QA testing. Engineers focus on building. GitHub and GitLab integration means no tool switching.
Product Teams
Product & Design TeamsSubmit requirements in plain language. Agents break them into stories, estimate, and plan. You review and approve. Ship faster with less overhead.
Teams That Care About Accountability
Compliance-Conscious TeamsEU-hosted or self-hosted. Every agent action logged and reversible. Built for teams that need to show what happened, why, and who approved it.
Built with Compliance in Mind
EU-hosted, auditable, and self-hostable. The technical foundation teams in regulated industries need to get started.
EU Data Residency
All data stored in European data centers. No US transfers. Self-host for full sovereignty over your infrastructure.
Tamper-Evident Audit Trail
Every agent action logged with full reasoning chain and SHA-256 hash chain. Designed for teams that need to demonstrate accountability.
Human Override on Everything
One-click rollback on any agent action. Confidence scoring surfaces uncertainty. Humans always have the final say.
Self-Host Option
Run the entire platform on your own infrastructure with Docker Compose. Bring your own LLM. Air-gap ready for sensitive environments.
Every plan includes AI agents. Enterprise adds governance.
Free forever for individuals. Upgrade for your team. Self-host on any plan.
Solo
Free foreverYou + AI agents. Everything you need for individual work and side projects. Not a trial — genuinely free forever.
- AI agents included
- Tamper-evident audit trail
- All core features
- Real-time collaboration
- Source control integration
Plus
RecommendedYour team + AI agents. Unlimited everything. Advanced testing and workflows.
- Everything in Solo
- AI agents with full autonomy controls
- Unlimited tasks, articles & members
- Advanced testing techniques
- Priority support
Enterprise
Coming nextSelf-hosted, SSO, pen-tested, SLAs. For teams in regulated industries that need full control over data and infrastructure.
- Everything in Plus
- Self-hosted deployment (Docker / Helm)
- Bring your own LLM (Ollama, vLLM, Azure OpenAI)
- SSO / SAML (MitID, NemLogin, Entra ID)
- Tamper-evident audit trail with verification API
Now on Mac, iPhone & iPad
Run your workspace natively on Apple devices. Join the TestFlight beta — one invite covers all three.
Latest from the blog
Deep dives and product updates from the Neuphlo team.
One format your team and your agents can read
We added support for Google's Open Knowledge Format. Here's why one open, shared format is how people in different departments and their AI agents finally stop re-explaining the same things.
How AI Agents Actually Handle Sprint Planning
"AI-powered" has become one of the most overloaded phrases in software. In most project management tools, it means autocomplete on task titles, or a button…
Bring Your Own LLM: Why It Matters More Than You Think
* * * When AI features became standard in SaaS tools, most vendors made the same quiet assumption: you'd use their model, through their API, on their…
Getting Started with Neuphlo
Everything you need to go from sign-up to your first working agent — setup, core concepts, and a guided walkthrough of the workspace.
Start free. Approve your first agent run in minutes.
Free forever for individuals. Every plan includes AI agents. EU-hosted or self-host — your data, your call.