[FR]: OpenCode Integration for AI-Driven Micro-Space Workflows #2381
Description
Is your feature request related to a problem? Please describe.
Currently, OpenCloud serves as a storage provider, but it lacks any built-in mechanism for structured, AI-assisted document processing workflows. Users who need to collect information through forms, attach files, and then have that information processed, reviewed, and refined through multiple stages are forced to rely on a patchwork of external tools (form builders, task managers, AI assistants, approval systems, version control) that do not integrate with their storage layer.
This means that a common scenario -- such as submitting an intake form with attachments, having AI extract and transform the relevant information, routing it through approval or review stages, and finally committing a finished result -- requires jumping between multiple disconnected platforms. There is no unified experience where the storage, the processing logic, and the review workflow live in the same environment. The result is fragmented data, lost context, and manual overhead that should not exist.
Describe the solution you'd like
Integrate oh-my-pi into OpenCloud to enable the creation of micro-spaces -- small, self-contained workspaces that combine structured forms, file storage, and AI-powered processing pipelines, all with Git-like version control.
The workflow would operate as follows:
1. Form Creation (with AI assistance)
A user creates a micro-space within OpenCloud. Inside that micro-space, they define a form (or have AI generate one based on a natural language description of what information is needed). The form can include text fields, structured data fields, and file upload slots.
2. Form Submission
A user (or multiple users) fills out the form, attaches relevant files, and submits. The submitted data and files live inside the micro-space as its initial state.
3. AI Processing Pipeline (Kanban-style stages)
The micro-space enters a processing pipeline that functions like a Kanban board. At each stage, oh-my-pi acts on the contents of the micro-space -- reading the form data and attached files, extracting information, transforming content, generating summaries, cross-referencing documents, or performing any other AI-driven task defined by the pipeline configuration.
Each stage transition is visible to the user. The micro-space moves from one column to the next (for example: "Submitted" to "AI Extraction" to "Review Needed" to "Approved" to "Complete"). At certain stages, the pipeline may pause and request human input -- an approval step, a request for additional information, or a manual review before proceeding.
4. Transparency and Interaction
The user can see the current stage of their micro-space at all times. They receive notifications when action is required from them (for example, when the AI flags missing information or when an approval gate is reached). They can inspect what oh-my-pi has done at each stage.
5. Git-like Commits for Micro-Spaces
Every change that oh-my-pi makes to the contents of a micro-space is recorded as a commit. The user can view a full history of edits -- who or what made each change, when, and why. They can diff between states, revert to a previous version, or branch a micro-space if needed. When oh-my-pi completes its work, the final state is committed as the canonical version, but the entire processing history is preserved.
This creates a system where the storage layer (OpenCloud), the intelligence layer (oh-my-pi), and the workflow layer (Kanban pipeline with approval gates) are fully integrated into a single coherent experience.
Describe alternatives you've considered
External form tools combined with automation platforms: Services like Typeform or Google Forms can collect data, and platforms like Zapier or n8n can trigger AI processing. However, the files and data then live across multiple systems, there is no unified version history, and the workflow visibility is poor. The user has to manually stitch together the pipeline.
Macro: Macro offers a similar concept of structured AI workflows with human-in-the-loop steps. However, it is a standalone platform and does not integrate with OpenCloud as a native storage backend. Using Macro would mean duplicating data outside of OpenCloud and losing the benefit of having everything in one place with OpenCloud's sharing, permissions, and storage management.
Git repositories with CI/CD pipelines: A technically sophisticated user could set up a Git repository where form submissions are committed as files, and CI/CD pipelines run AI scripts at each stage. This approximates the desired behavior but requires significant engineering effort, offers a poor user experience for non-technical users, and does not integrate with OpenCloud natively.
Manual processing: Users can simply store files in OpenCloud and process them manually or with standalone AI tools. This works but does not scale, provides no audit trail, and defeats the purpose of having an intelligent storage platform.
Additional context
The core idea is that OpenCloud should evolve beyond passive storage into an environment where data can be actively worked on by AI agents in a structured, transparent, and version-controlled manner. The micro-space concept keeps these workflows contained and manageable -- each one is a small, focused unit of work rather than a sprawling project.
The Kanban-style visualization is important because it gives users immediate understanding of where their submission stands and what is happening to it without requiring them to understand the underlying AI processing. The Git-like commit model is important because it provides full auditability and reversibility, which are critical for any workflow that involves automated edits to user data.
oh-my-pi already provides the AI capability. OpenCloud already provides the storage and collaboration layer. This feature request is about creating the connective tissue between them -- the micro-space as the unit of work, the pipeline as the processing logic, and the commit history as the trust and transparency mechanism.
For reference, Macro demonstrates many of these concepts in a standalone product, particularly around structured AI workflows with human review gates. The proposal here is to bring similar capabilities natively into the OpenCloud ecosystem by leveraging oh-my-pi as the processing engine and OpenCloud spaces as the storage foundation.