Week of 23 Feb 2026¶

What happened¶

The discussion focuses on progressing the Machine World project and its associated initiatives, Marga Sakaccha and Inner Realm. The team outlines immediate priorities, beginning with the validation of the current demo website container before releasing it publicly. A major technical goal is establishing an automated media pipeline that transforms Dhamma audio recordings into animated videos using static images, time frames, and containerized workflows. The lead speaker introduces the integration of AI agents equipped with Model Context Protocol (MCP) tools and agentic skills to handle audio cleaning, visual generation, and automated distribution to platforms like YouTube and podcasts. On the Dhamma side, the team explores visual representations of Abhidhamma concepts, specifically Citta, Citta Vithi, and [doctrine term] causal relations showing past and future Vipaka. Validation of scriptural accuracy is emphasized, with suggestions to consult monks like [name withheld] and use OCR or LLM tools (such as Gemini) for text verification. Responsibilities are coordinated among team members, including [name withheld] for pipeline skills, [name withheld] for Dhamma video curation, and [names withheld] for development and data validation.

Key points¶

Decided · The first milestone is validating the current demo website container before making it publicly accessible.
Said to exist · Connecting or pushing to the repo automatically triggers updates on the live demo site.
Planned · Expose the validated website data via an MCP tool interface so external voice models and LLMs can accurately query and respond on Abhidhamma topics.
Planned · Once validated, the website content will be published to the public website.
Planned · The third milestone is developing visual representation components for Abhidhamma Citta and [doctrine term] causal relationships.
Planned · Speaker 1 will share resources and links regarding Claude/agent skills and MCP with [name withheld].

Ideas and open questions¶

Idea · Consult [name withheld] to review and verify the complex Abhidhamma concepts modeled in the system.
Idea · Utilize AI models supporting Sinhala, such as Gemini, along with OCR to assist in cross-checking and validating Dhamma text data against books.
Open · Whether 3D plane animations and smooth image zoom/diffusion transitions can be achieved using tools like CapCut.

All details (4 topics, 3 more points)

Demo Site Validation and Public Release Workflow: The team reviews the current demo site deployment linking Machine World, Marga Sakaccha, and Inner Realm. They establish a milestone roadmap: first validate the containerized demo website content, then publish it publicly, and subsequently build out visual components.
Abhidhamma Visualization: Citta, Vipaka, and [doctrine term]: Participants discuss designing a graph-based visual model to depict how past karmic results (Vipaka) arise, how present consciousness functions, and how future states emerge according to [doctrine term] relations. They plan to include Citta Vithi (cognitive processes) and emphasize scriptural accuracy.
Automated Media Pipeline via MCP and Agent Skills: The team discusses automating video generation from historical audio sermons (e.g., Damme Kanda Dhammaratana Thero) and custom meditation clips. The architecture leverages lightweight containers, scripts, Model Context Protocol (MCP), and agent skills to clean audio, sync images to timeframes, and auto-upload to YouTube and podcast channels.
AI Integration, MCP Exposure, and Data Verification: Plans are outlined to expose validated Marga Sakaccha web data through MCP tools to allow LLMs and voice models to interactively answer Dhamma queries. Methods for content verification using OCR, Sinhala-capable LLMs like Gemini, and monastic consultation are proposed.
Planned · Generated video and audio content will be automatically distributed to YouTube, Facebook, social platforms, and podcast feeds.
Idea · The automated agent architecture will combine agent skills and MCP tools to manage audio processing and distribution.
Idea · The audio-to-video pipeline will use containerized execution where an ephemeral container spins up on API trigger, renders the video, and is destroyed.