[Split Scope Migration](2) Remind agents to read split-scope on planning prompts

[EdbertChan]

Summary

Today product/skills/split-scope only loads when the model decides to pick it
from its description. This workflow adds a catstack hook,
engine/hooks/split-scope/, that injects a short reminder to read the skill
whenever a prompt plans a PR stack, several PRs, a break-up into slices, or a
migration. It copies the shape of engine/hooks/build-the-lever/: a regex
detector, fail-open entrypoints for Claude, Codex, and Cursor, installers,
and tests that pin when it fires and when it stays silent.
Gated on workflow 1 (the split-scope section) merging.

Test Plan

Test Plan

• python3 -m unittest discover -s engine/hooks/split-scope/tests -v && python3 scripts/check_hook_test_coverage.py engine/hooks/split-scope && python3 -m unittest discover -s tests -p "test_install.py" — Review claim:

• The hook's own tests, the hook-coverage gate, and the install tests all pass.
  Review lane:

• proof
  Safety invariant:

• Proof-only; this read-only check changes no files.
  Effectiveness measurement:

• The command below exits 0.
  Slice rationale:

• One proof step for the one hook added.
  Architectural effect:

• None; verification only.
  Goal:

• Prove the hook fires and stays silent as specified and installs cleanly.
  Motivation:

• The tests run the real entrypoint, so a pass means the reminder really reaches the agent.
  Alternative considerations:

• Option A (chosen): run the hook tests, the coverage gate, and the install tests together.
  Implementation details:

• Run the three checks in order.
  Non-goals:

• No product edits here.
  Layer: app_regression
  Feature state: active

• bash scripts/scrub-handoff-artifacts.sh — Review claim:

• No ephemeral handoff files are left behind before the PR merge gate.
  Review lane:

• proof
  Safety invariant:

• This read-only check changes no files.
  Effectiveness measurement:

• bash scripts/scrub-handoff-artifacts.sh exits 0.
  Slice rationale:

• The scrub check runs last and stays separate from implementation.
  Architectural effect:

• No architecture change.
  Goal:

• Confirm no handoff artifacts remain.
  Motivation:

• Uncommitted handoff files do not flow across worktrees.
  Alternative considerations:

• Option A (chosen): a read-only absence check.
  Implementation details:

• Run the handoff scrub script without --apply.
  Non-goals:

• Do not modify source files.
  Layer: e2e_regression
  Feature state: active

Revert Plan

Revert Plan

• Safe to revert? Yes
• Revert command: git revert <sha>
• Post-revert steps: None
• Data migration? No

---

Note

Low Risk
Read-only, fail-open prompt injection with no auth, data, or product behavior changes; risk is limited to occasional extra context in agent sessions.

Overview
Adds a new split-scope catstack hook (engine/hooks/split-scope/) that injects a short reminder to read the existing product/skills/split-scope skill when a user prompt looks like multi-slice / multi-PR planning (PR stacks, “split this into…”, migration plans, etc.), instead of relying on the model to pick the skill from its description alone.

The implementation mirrors build-the-lever: shared regex detection in detect.py, fail-open Claude/Cursor/Codex prompt hooks (Cursor defers inject to postToolUse), installers, hook JSON fragments, and unit tests that pin fire vs. silent cases. install.sh and tests/test_install.py wire the hook into Claude, Cursor, and Codex installs.

^{Reviewed by Cursor Bugbot for commit fceccee. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_413c7c7e-6a6d-427e-bf15-2e79873b136c)