Using AWS to Manage AI Spec-Driven Development Artifacts

说明与用途

利用 S3、S3 Vectors 和 Bedrock 快速存储、索引和语义搜索 AI 代理生成的 Markdown 产物，实现即时一致性、章节级搜索和跨团队共享。

问题

Managing the large number of markdown artifacts produced by AI agents in spec‑driven development becomes unwieldy, pollutes the code base, and hinders sharing and search.

结果

The Arkeology project delivers immediate consistency, semantic and section‑level search, cross‑team sharing, rich metadata, commit traceability, a visual browser, and full artifact lifecycle management using only three AWS services.

工作流

• Store markdown artifact in S3
• Generate H2 section embeddings with Bedrock Titan Text Embeddings v2
• Store embeddings in S3 Vectors with metadata
• Annotate S3 object with source, relationships, commit SHA
• Expose MCP server tools for store, manage, search
• Use MCP App to browse artifacts

证据

Doing spec-driven development with AI agents, I quickly ran into a problem: managing all the artifacts they produce.

来源

The basic approach is to store them in a folder in the repo (e.g. docs/), with an index.md file listing each file with a short description of its content.

来源

S3 stores all the markdown artifacts

来源

Combining just these 3 AWS services gave birth to the Arkeology project with the following features: Immediate write-then-read consistency, Semantic search with metadata filtering, Section-level search precision, Cross-team knowledge sharing, Rich, filterable metadata, Commit-to-artifact traceability, Arkeology Studio, Full artifact lifecycle, Migration

来源