# AI use cases for Stru Info (tntexploding/StruInfo)

Repo: [tntexploding/StruInfo](https://github.com/tntexploding/StruInfo) | Use cases: 4 | Page: https://dedupedai.com/use-cases/tntexploding-struinfo/

StruInfo is a self-hosted personal document workspace that turns feeds and uploaded files into searchable, traceable notes with optional AI review, Q&A, and reports.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Run a private knowledge base on your own server: https://dedupedai.com/use-cases/tntexploding-struinfo/self-host-personal-knowledge-base/
- Review RSS and Atom articles into a searchable archive: https://dedupedai.com/use-cases/tntexploding-struinfo/review-rss-feeds-into-searchable-notes/
- Produce a weekly or monthly report from your sources: https://dedupedai.com/use-cases/tntexploding-struinfo/generate-periodic-reports-from-sources/
- Let an external AI agent file documents through MCP: https://dedupedai.com/use-cases/tntexploding-struinfo/let-an-agent-file-documents-over-mcp/

## How to run a private knowledge base on your own server with Stru Info

Repo: [tntexploding/StruInfo](https://github.com/tntexploding/StruInfo) | Difficulty: Intermediate | Time: 60 minutes | Basis: From the README

Install Stru Info with Docker Compose and PostgreSQL 18 as a single-owner instance.

**Who it is for:** A solo researcher who wants their reading and documents kept on a machine they control, instead of a hosted notes service.

**The problem:** Reading material is scattered across feeds, PDFs, and web pages, and there is no single private place to search and trace it.

**The outcome:** A running single-owner Stru Info instance with its database and data directory kept separate from the app image.

### Steps

1. Follow the first-install guide in docs/getting-started.md to set up the database role, configuration, and migrations.
2. Start the app with `docker compose -f compose.production.yaml up -d app`.
3. Check that the container is running with `docker compose -f compose.production.yaml ps`.
4. Open Stru Info in a browser on the host port set by STRUIINFO_HTTP_PORT, keeping the port on loopback or a trusted private network.

### Prompt to paste into your agent

```text
In the repo tntexploding/StruInfo, follow docs/getting-started.md to deploy a single-owner instance at <host-path> using Docker Compose and PostgreSQL 18. Report the output of `docker compose -f compose.production.yaml ps` and confirm the app container is running.
```

### Caveat

The README says the app has no multi-user login, so the port must stay on loopback or a trusted network, or be reached through an authenticated HTTPS proxy, VPN, or SSH tunnel.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/tntexploding-struinfo/self-host-personal-knowledge-base/

## How to review RSS and Atom articles into a searchable archive with Stru Info

Repo: [tntexploding/StruInfo](https://github.com/tntexploding/StruInfo) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README

Subscribe to feeds, accept the articles you want, and search them in one place.

**Who it is for:** A researcher who follows several blogs or newsletters and wants only the useful articles kept, with a link back to each original.

**The problem:** Feeds produce more articles than anyone can read, and saved links are hard to search later.

**The outcome:** A set of accepted, confirmed articles that you can search and cite, each linked to its source.

### Steps

1. In the 文档入口 page, add an RSS or Atom source and read the incoming article text.
2. Accept or reject each article, or let AI review it. Leave uncertain items waiting for your decision.
3. In the 工作台, check the segments, tags, and links, then click 确认入库 for each accepted public document.
4. In the 数据阅览 page, search the confirmed content by body text or tag.

### Prompt to paste into your agent

```text
In the repo tntexploding/StruInfo, add the RSS or Atom feed at <feed-url> as a source in the 文档入口 page and review its incoming articles. Report how many were accepted, rejected, or left waiting, and verify that a search in 数据阅览 returns a confirmed article with its source link.
```

### Caveat

Confirmed content only becomes available to search after you confirm it in the 工作台, so an accepted article that is not yet confirmed will not appear in search.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/tntexploding-struinfo/review-rss-feeds-into-searchable-notes/

## How to produce a weekly or monthly report from your sources with Stru Info

Repo: [tntexploding/StruInfo](https://github.com/tntexploding/StruInfo) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README

Pick a time range and topic in the report page to get a Markdown or printable PDF summary.

**Who it is for:** A analyst who must brief a team every week or month on a set of topics drawn from their saved sources.

**The problem:** Writing a periodic summary by hand means rereading many saved items and tracking their sources manually.

**The outcome:** A weekly, monthly, or topic report built from confirmed material, with source citations to check.

### Steps

1. Confirm the sources you want included in the 工作台 so they are available to reports.
2. Open the 报告 page and choose the time range and topic.
3. Generate the report, then review its citations against the original sources.
4. Download the Markdown file or use the browser print option to save it as a PDF.

### Prompt to paste into your agent

```text
In the repo tntexploding/StruInfo, use the 报告 page to generate a monthly report for <topic> covering <month>. Return the Markdown output and check that each claim includes a source citation that can be traced back to a confirmed document.
```

### Caveat

The README notes that AI answers and reports still need checking against the sources, and it does not include any built-in external push channel for delivering reports.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/tntexploding-struinfo/generate-periodic-reports-from-sources/

## How to let an external AI agent file documents through MCP with Stru Info

Repo: [tntexploding/StruInfo](https://github.com/tntexploding/StruInfo) | Difficulty: Advanced | Time: 90 minutes | Basis: From the README

Connect an agent with a scoped token so it can import, split, tag, and link your documents.

**Who it is for:** A developer or researcher who already works with an AI agent and wants it to manage the knowledge base without using the web interface.

**The problem:** Filing each document by hand in the web interface takes too long when the work is repetitive.

**The outcome:** An agent that reads or writes Stru Info content through the MCP interface, limited to the permissions you granted.

### Steps

1. Read docs/agent-interface.md and create an agent token with read-only access, or with import, split, tag, and link write access.
2. Configure your MCP-capable agent client to point at the Stru Info MCP endpoint described in docs/agent-interface.md. (inferred, check the README)
3. Ask the agent to list available documents before it writes anything.
4. Review the agent's changes in the 工作台 and confirm them with 确认入库.

### Prompt to paste into your agent

```text
In the repo tntexploding/StruInfo, use the MCP interface described in docs/agent-interface.md with a read-only token to list documents matching <topic>. Return the titles and IDs, and verify that no write operation was performed.
```

### Caveat

Agent write access can add or change items, so keep the token scoped and review agent edits in the 工作台 before confirming them.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/tntexploding-struinfo/let-an-agent-file-documents-over-mcp/