OWASP / OWASP/www-project-api-security-testing-framework
Decision needed: fund API credits to verify crAPI chatbot compliance (rows 16-18), or accept as permanent Partial
Nobody has claimed this yet.
- Dominant language
- Java
- Stars
- 150
- Forks
- 41
- Avg merge
- 2d 5h
- Merged PRs (30d)
- 3
Description
Tracked in docs/TRACEABILITY.md, crAPI rows 16-18.
#90's prompt-injection check correctly reaches crAPI's real chatbot endpoint and fails gracefully without a live OpenAI key — mechanics verified, actual compliance never exercised (no API credits spent).
Per the scope-boundary note added in #113: this challenge is really two separable things — (a) reaching the endpoint and sending a crafted prompt (in scope, already works), and (b) judging whether the model semantically complied with the injected instruction (closer to LLM safety/behavioral evaluation than structural API testing).
This isn't a code task by itself — it needs a maintainer decision on whether to fund real API credits to attempt (b), or accept these three rows as a permanent "Partial" given the scope-boundary nuance. Not blocking other work either way.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with docs/TRACEABILITY.md rows 16-18, then review #90’s prompt-injection check and the scope-boundary note in #113. Confirm the endpoint mechanics and the missing semantic-compliance evaluation; done means the maintainer decision—fund credits for (b) or accept Permanent Partial—is recorded.
Written by the indexing model from the issue text.
Assessment
- Domain
- documentation, security
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Clearly specified
- Newbie friendliness
- 25/100