[EdbertChan]

Summary

When a task produces something a person will read, the agent often builds the data collector first. The reader then sees the result last and asks for changes.

This adds one step to the planning guide. Build a mock on fake data first. Agree on what is shown and what each number means.

Only then plan the real data work.

Review Claim

Approve a planning step that asks for an approved mock before any data collection is planned, whenever the output is meant to be read.

Review Lane

behavior

Review Unit

product-skill

Safety Invariant

It adds guidance text and one example. No code runs differently, and removing the paragraph restores the old guide.

Slice Rationale

This is a separate claim from the Invoker gate and from the hand-back rule, and it lives in a different review unit.

Non-goals

• No change to how other plans are made.
• No change to the Invoker gate; that ships separately.

Measured

none: this adds guidance text; there is no number to measure

Test Plan

Test Plan

• python3 scripts/ci/check_skill_test_coverage.py --base origin/main --head HEAD printed ok skill test coverage
• python3 scripts/ci/check_no_dated_provenance.py --base origin/main printed ok no dated provenance
• python3 scripts/ci/check_skills_three_harnesses.py, check_ecosystem_boundaries.py, check_skill_file_refs.py, check_skill_trigger_policy.py and check_skill_trigger_mechanism.py each exited 0
• python3 engine/skills/make-pr/scripts/preflight.py --base origin/main --dry-run printed unit product-skill: 2 file(s) and ok preflight passed
• Not run: a live session where the skill fires on a real request. Blocker: the skill only loads after this merges and is installed

Revert Plan

Revert Plan

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

🤖 Generated with Claude Code

[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: 701c06bc-f186-45a7-82f1-82e764059720)

[mergify]

Queued — the merge queue status continues in this comment ↓.

[EdbertChan]

@Mergifyio queue

[mergify]

Merge Queue Status

• ✅ Entered queue — 2026-10-12 20:04 UTC · Rule: admin-bypass · triggered by @EdbertChan with the @mergifyio queue command
• ❌ Checks failed — 2026-10-12 20:24 UTC · on draft merge queue: checking #1332 + #1354 + #1358 + #1359 together on main (2f93d6c) #1361
• ✂️ Bisecting to identify the failing PR (round 1/1)
• ✅ Checks passed · on draft merge queue: checking #1358 on main (2f93d6c), stacked on #1332 and #1354 #1365
• ✅ Merged — 2026-10-12 22:58 UTC · at afd48c4ab6c1b77b3c3a3f504b3c34ad0ea645e8 · squash

This pull request spent 2 hours 54 minutes 1 second in the queue, including 1 hour 42 minutes 19 seconds running CI.

Required conditions to merge

• check-success = lint
• check-success = test
• check-success = validate