Show HN: Selvedge – an append-only log of what your AI agent already rejected

Description and use

Here's a brief summary of the open-source project Selvedge:

Selvedge is an append-only log of AI agent rejections, allowing developers to add AI capabilities without starting from a blank model stack. This project is particularly useful for prototyping AI features, building Retrospective Active Learning (RAG) or agent workflows, and evaluating model tooling. While it has medium production readiness, it requires manual inspection and verification of its quality signals, license, maintenance, documentation, and release cadence before adoption.

The value proposition of Selvedge lies in its ability to help developers skip the time-consuming process of building a model stack from scratch, allowing them to focus on more advanced AI capabilities. The practical adoption path involves manually inspecting the project's metadata, verifying its quality signals, and ensuring that it meets the necessary standards for production use.

Production readiness is classified as medium, indicating that while Selvedge is useful for prototypes or internal workflows, it requires careful evaluation and testing before being deployed in a production environment. This is due to the limited quality signals available and the need for manual inspection and verification.

Use in a workflow

Show HN: Selvedge – an append-only log of what your AI agent already rejected helps add AI capability without starting from a blank model stack.

• prototype AI features
• build RAG or agent workflows
• evaluate model tooling

Limitations and review notes

Medium: useful for prototypes or internal workflows, with dependency and maintenance checks before production.