Add explicit concept-set approve/edit/rerun loop for ACP-based Keeper workflow
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 12
- Forks
- 7
- Avg merge
- 2m
- Merged PRs (30d)
- 12
Description
Title
Add explicit approve/edit/rerun loop for ACP-generated Keeper concept sets
Body
Problem
The ACP-based Keeper flow can now generate, reuse, and overwrite approved concept-set artifacts, but there is still no explicit human approval/edit/rerun loop in the workflow.
Manual edits to concept-sets-approved/*.json are consumable, but editing remains out-of-band.
Current behavior
- generated concept sets can seed approved concept sets
- approved concept sets can be reused
- approved concept sets can be overwritten from generated output
- there is no first-class approval state machine or in-shell review loop
Risk
- workflow remains operationally awkward
- users may accidentally rerun with stale or unreviewed concept sets
- approval provenance is weak
Requested change
Add an explicit approval/edit/rerun concept-set workflow for ACP-based Keeper.
Acceptance criteria
- users can clearly choose between:
- review current generated concept sets
- keep approved concept sets
- overwrite approved from generated
- rerun concept-set generation
- approval provenance is persisted in workflow state
- prompts/status make current approved/generated sources clear
- regression tests cover reuse, overwrite, and rerun paths
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
Start by locating the ACP-based Keeper workflow and the handling of concept-sets-approved/*.json. Trace how generated and approved concept sets are selected, reused, and overwritten, then identify where workflow state and prompts are maintained. Done means the review, keep, overwrite, and rerun choices are explicit, provenance is persisted, source status is clear, and regression tests cover reuse, overwrite, and rerun paths.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- cli
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100