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

How to flag risky behavior in AI traces with Span-01 with Awesome Decision Models

Label each AI trace for behaviors you define, such as prompt injection or agent loops.

Credentials. The OpenRouter key grants access to your account, so keep it in an environment variable and do not commit it to the trace review repo.

Who it is forAn AI product engineer reviewing agent traces who needs to find prompt injection, hallucination, or loops without reading every trace by hand.

The problemReading traces manually does not scale, and keyword search misses many behaviors described in plain language.

The outcomeEach trace gets present, absent, or not observable for every behavior you define, with a list of traces to review first.

What you need first

  • A web browser to open the repo's README and the linked provider docs
  • An account and API key for the hosted provider you choose (for example TypeSafe console, OpenRouter, or Microsoft Foundry)
  • For local runs: Ollama installed, or a Python environment with the package named in the model's own page
  • A list of your own options (labels, tool names, or rubric levels) to score

Steps

  1. Open the Hosted APIs section and read the Span-01 entry, which classifies AI traces for behaviors you define in plain language.
  2. Get an OpenRouter API key and find the Span-01 page at openrouter.ai/respan/span-01 linked from the README.
  3. Write one plain-language definition for each behavior you want to flag, such as prompt injection, hallucination, or agent loops.
  4. Run the classifier over a sample of traces and save each result with its trace ID.
  5. Review every trace marked present, and spot-check a sample marked absent.

Prompt for your agent

Using the Span-01 entry in github.com/AnotiaWang/awesome-decision-models, classify the traces in <traces-folder> for prompt injection and agent loops through the OpenRouter route the README links. Expected output: one present, absent, or not observable value per behavior per trace. Verify by reading three traces marked present and confirming the behavior is actually there.

Caveat. The README gives Span-01 no accuracy numbers, so check its labels against traces you have already reviewed before trusting it on a larger set.

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