crate / crate/about

Prompt: LLMs emit URLs that do not exist

Open
#54 1 comment 0 reactions 0 assignees View on GitHub
bug
Dominant language
Python
Stars
1
Forks
0
PR merge metrics
No merged PRs in 30d

Description

## Problem

LLMs hallucinate, in this case about URLs. The question was:

> "CrateDB does not seem to provide an AUTOINCREMENT feature?"

## Details

Both text fragments include URLs that don't exist, or are wrong. The correct URL would be https://cratedb.com/docs/guide/performance/inserts/sequences.html.

> CrateDB does not offer automatic incrementing IDs due to distributed scaling concerns. Use UUIDs, timestamps, or an application-managed sequence as needed. See the [CrateDB documentation: Autogenerated sequences and PRIMARY KEY values](https://cratedb.com/docs/crate/howtos/en/latest/data-modeling/sequences.html) for examples and best practices.

> [CrateDB: Autogenerated sequences and PRIMARY KEY values](https://cratedb.com/docs/crate/howtos/en/latest/getting-started/autogenerated-sequences.html)

> [CrateDB Community: Sequences in distributed databases](https://community.cratedb.com/t/how-to-work-with-primary-key-auto-increment-in-cratedb/957)

## Reproduce

The command was:

```sh
uvx llm --model gpt-4.1 \
--fragment https://cdn.crate.io/about/v1/llms-full.txt \
--system-fragment https://cdn.crate.io/about/v1/instructions-general.md \
"CrateDB does not seem to provide an AUTOINCREMENT feature?"
```

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by running the provided uvx llm command with the listed model and fragment URLs, then inspect the source behind https://cdn.crate.io/about/v1/llms-full.txt and its instructions fragment. Trace where the incorrect CrateDB links are supplied or generated. Done means the response points to https://cratedb.com/docs/guide/performance/inserts/sequences.html rather than the nonexistent URLs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai, content, documentation
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.