Plan CEO Review Skill | Agent Skills

Plan CEO Review

Role

Review a planning artifact for product value and scope discipline before execution begins. Run this as an isolated plan-review boundary and merge back only the verdict summary and required changes.

This skill does not rewrite the whole plan by default. It produces a decision verdict that upstream planning stages must respect.

Use When

a plan looks complete but may still be too large
the package needs a value or timing judgment
scope may need to be reduced before implementation

Inputs

one planning artifact at a time:
- PRODUCT_INTENT.md
- PRD.md
- PLAN.md
optional supporting context:
- ASSUMPTIONS.md
- BLOCKERS.md
- SPEC.md

Review Questions

Why should this work happen now?
What is explicitly out of scope?
Is the likely implementation cost justified by the user value?
Should scope be reduced, held, or rejected before execution?

Applied Rubric

require a clear why-now argument, not only a complete artifact
require explicit non-goals and a core success state
prefer narrowing to the highest-value user path when value is uncertain
treat observability, rollout risk, and support burden as part of scope cost
if value is weak even after narrowing, return hold_scope or fail

Verdict Contract

Return exactly one of:

pass
conditional_pass
scope_reduction
hold_scope
fail

Output Shape

planCeoReview:
  artifact: "PLAN.md"
  verdict: "scope_reduction"
  summary: "Current phase list is broader than the near-term value case."
  requiredChanges:
    - "split admin analytics into a later slice"
  assumptions:
    - "Current run targets only core user path"
  blockers: []

Rules

prefer scope reduction over speculative expansion
do not treat completeness as value
if value is weak or timing is poor, use hold_scope or fail
if details are missing but not critical, use conditional_pass

References

skills/product-orchestrator/SKILL.md
skills/moonshot-plan-writer/SKILL.md

Agent Skills: Plan CEO Review

Install this agent skill to your local

Skill Files