DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to audit a clinical chat assistant's response to missing evidence with Keystone Bench

Run the quick set to see whether a model's medical advice moves when a decisive fact changes.

  • Top pick
  • Research
  • Intermediate
  • 1 hour
  • From the README
  • Any agent

Credentials. The run needs an API key for an OpenAI-compatible endpoint, so store it in an environment variable and do not commit it.

Who it is forAn AI safety or clinical evaluation researcher who needs to test a chat assistant before it answers patient-style questions.

The problemA physician-written rubric scores a reply to the message the physician saw, so it says nothing about what the assistant does when a decisive fact is missing, replaced, or contradicted.

The outcomeA per-family report with rates for how often the assistant's action follows the evidence on the twins, with the negative controls reported separately.

What you need first

  • Python with pip to install keystone-bench from PyPI
  • An API key for an OpenAI-compatible endpoint, such as OpenRouter, to run models and a judge from a different vendor
  • Network access so that keystone build can fetch HealthBench from OpenAI's public copy and the label files
  • Enough disk space and time for the local rebuild of the twins, since the release ships span references rather than benchmark prose

Steps

  1. pip install keystone-bench
  2. keystone build
  3. keystone estimate --family all --rubric
  4. export OPENROUTER_API_KEY=...
  5. keystone run --family all --layer quick --model openrouter/openai/gpt-5.6-terra --judge openrouter/anthropic/claude-sonnet-5

Prompt for your agent

Use the keystone-bench repo to run the quick layer against <target model> with a judge from a different vendor, using keystone run --family all --layer quick. Confirm that runs/<run>/REPORT.md exists and check the per-family table, including whether controls pass 35 percent of the run.

Caveat. Results describe the assistant on these twins only, and the action judge must come from a different vendor than the model under test.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page