spike-probelisted

# this skill Rapid spike harness. Probe the goal in the working tree with the lightest verification that proves the path, exit through the project's fast quality gates, stash everything with a recovery ref, and hand back a neat commit-by-commit plan — optionally replaying it immediately. A probe **mutates the working tree freely** — mutation is the point; the code is the sensing instrument. What it never touches is history: like an attached kprobe, it patches the live system to take its measurement and detaches without a trace in `git log`. This is a slash command, not a model-invocable skill: it mutates the working tree and (in `--replay`) creates commits, so it must be user-explicit, not router-inferred. ## Core thesis A spike answers "does this path work?" — it is **evidence, not a deliverable**. The probe's value survives as (a) a stash you can reapply and (b) decisions the plan records. Committing spike code converts unreviewed exploration into history. Three disciplines: 1. **Zero commits during the spike.** The probe ends in a stash and a plan, never in `git commit`. 2. **Right-sized verification.** During the probe, run only what proves the path (a smoke test, a scoped test run). At spike exit, run the project's fast gates once so the plan starts from known-green knowledge. Defer what CI covers to CI. 3. **The plan is the product.** Every probe ends with a numbered commit sequence mapping stash contents to commits, with the discovered gate com