Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →hashgraph-online/worktree
Use for non-trivial or parallel work that should happen in a separate git worktree.
Passed 15 of 18 test cases with the skill, and 9 of 18 without it.
and 5 more in the full report
7 fixed11 passed either way4 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
We are preparing a multi-repo change for order-service and inventory-service. Tests in order-service pass, but inventory-service has 1 failing test. Can we create the worktree for order-service now?
| Checked | Without skill | With skill |
|---|---|---|
| halts multi-repository worktree setup if any baseline is failing | Missed | Passed |
The instructions your model reads when the skill loads.
If topology=multi-module (triage announcement or coordinator spec/plan declaration), read ../references/multi-module.md. Treat each repository as an independent Git and worktree boundary; never run a workspace-wide Git mutation from their common parent.
Before creating worktrees, resolve each registered repository and verify status and baseline. Map paths in the coordinator plan and use the change-set ID in names. A missing repo, unrelated dirty change, or red baseline blocks the change. Detached HEAD is allowed for read-only integration, not implementation. Do not clone or clean automatically.
git worktree add ../<repo>-<topic> -b <type>/<topic> # <type>: feat | fix | chore | ...
cd ../<repo>-<topic> && <setup> && <run tests>
Baseline must be green in every participating repository. If red, STOP - don't build on broken ground. Done → ship.
Worktrees are separate directories from the main repo. Semantic indexing (code graph, symbol search) covers the original repo, not the worktree. File tools (read, grep, glob) work normally. For short-lived worktrees this is acceptable; for long sessions, note this limitation to the agent at dispatch.
test cases passed, vs 9 / 18 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.