leanprover-community / leanprover-community/iris-lean
Debug mode for tactics
Nobody has claimed this yet.
- Dominant language
- Lean
- Stars
- 218
- Forks
- 63
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 22
Description
During the Iris workshop, it was brought up that if someone wants to apply LLMs to help with Iris proofs, it would be beneficial to have a verbose output mode. I haven't thought about the design, so we can discuss it here. Perhaps, one could add a flag that in case of errors makes tactics dump the Iris goal in a more structured way. Plus, it would be nice if the output can be used to state intermediate goals (that is, output contains a valid Entails statement).
If anyone has more ideas on what would be helpful, feel free to propose other ideas. I haven't used LLMs for any big scale things, so I am not sure what would be helpful or not.
Possibly, there can be also some structured summary of Iris tactics?
Contributor guide
No contributing guide indexed for this repository
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
No files, tests, or entry points are named. Start by reviewing the existing Iris tactic output and discussing the proposed debug-mode goals, including structured Iris goals and valid Entails statements. Done means the design is agreed and the resulting output requirements are clearly testable.
Written by the indexing model from the issue text.
Assessment
- Domain
- tooling
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100