docs: enhance ARCHITECTURE.md with comprehensive pipeline documentation

[hoangsonww]

Expand from 1,024 to 1,441 lines with 25 Mermaid diagrams covering:

• Agentic Coding Pipeline: orchestration flow, class hierarchy, agent table, state keys, task resolution (GitHub/Jira/text), streaming API spec
• Agentic RAG Pipeline: full multi-agent flow with subtask loop, agent details (model/temp/tokens), Evidence model, FAISS internals, intent classification
• Social Media Automation: 12-tool registry, campaign creation sequence diagram, scheduler DB schema with indexes, background service loop
• Cross-Pipeline Integration: shared infra diagram, complete state key summary table across all 4 pipelines, error handling patterns

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly expands the ARCHITECTURE.md file, providing a much more comprehensive and detailed guide to the system's design. The update focuses on clarifying the internal workings, data flows, and interconnections of the various agentic pipelines, making the overall architecture more understandable and maintainable for developers. It introduces numerous Mermaid diagrams, tables, and descriptive text to illustrate complex processes and components.

Highlights

• Agentic Coding Pipeline Documentation: Expanded documentation for the Agentic Coding Pipeline, including detailed orchestration flow, agent class hierarchy, state management, task resolution, and streaming API specifications.
• Agentic RAG Pipeline Documentation: Enhanced the Agentic RAG Pipeline documentation with a comprehensive multi-agent orchestration flow, agent details, evidence model, intent classification, streaming API, and FAISS vector store specifics.
• Social Media Automation Documentation: Introduced detailed documentation for the Social Media Automation, covering its architecture, a 12-tool registry, campaign creation workflow, scheduler database schema, and background service operation.
• Cross-Pipeline Integration and Error Handling: Added a new section on Cross-Pipeline Integration, illustrating shared infrastructure, providing a summary of state keys across all pipelines, and outlining a consistent error handling pattern.
• Agentic Data Pipeline Overview: Incorporated a new section for the Agentic Data Pipeline, detailing its location, exposure, and state.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request significantly enhances the ARCHITECTURE.md documentation by adding detailed explanations and Mermaid diagrams for various pipelines, including the Agentic Coding, RAG, and Social Media Automation pipelines. The changes are comprehensive and greatly improve the understanding of the system's architecture. My review focuses on ensuring the consistency and accuracy of the newly added information. I've identified a few minor discrepancies in the documentation regarding agent inputs, tool counts, and data types that could be clarified.

[gemini-code-assist]

The Pipeline State and State Key Summary tables correctly list a feedback key for retry loops. However, this table doesn't show feedback as an input to the CodingAgent. To improve the code on retries, the agent should receive the feedback from the failed test or QA step. Please consider updating the Input for CodingAgent to include feedback.

Suggested change

	| `CodingAgent("gpt-coder")` | OpenAI GPT-4o | Generate/improve code | `task`, `proposed_code?` | `proposed_code` |
	| `CodingAgent("claude-coder")` | Claude | Generate/improve code | `task`, `proposed_code?` | `proposed_code` |
	| `CodingAgent("gpt-coder")` | OpenAI GPT-4o | Generate/improve code | `task`, `proposed_code?`, `feedback?` | `proposed_code` |
	| `CodingAgent("claude-coder")` | Claude | Generate/improve code | `task`, `proposed_code?`, `feedback?` | `proposed_code` |

[gemini-code-assist]

There's a discrepancy in the number of tools for the SocialMediaAgent. The diagram here mentions 12 tools (5 social, 5 content, 2 utility), but the Tool Registry table below lists only 10 tools (5 platform/social, 5 content). The code in src/agentic_ai/agents/social_media_agent.py also seems to initialize only 10 tools. Please clarify if there are 2 missing utility tools and update the table, or if the diagram should be updated to reflect 10 tools.

[gemini-code-assist]

The type datetime is not a native JSON type. For clarity and consistency with the API and database schema (which uses TEXT), it would be better to specify the type as string and mention the format, for example string (ISO 8601).

Suggested change

	| | `scheduled_time` | `datetime` | When to publish |
	| | `scheduled_time` | `string (ISO 8601)` | When to publish |