| name | review |
| description | Run Meridian's 7-question review checklist on an epic or plan |
| user-invocable | true |
| context | fork |
| agent | meridian-reviewer |
| argument-hint | [path to epic or plan document] |
| allowed-tools | Read, Glob, Grep |
Review this document: $ARGUMENTS
Read the document and thestudioarc/personas/meridian-review-checklist.md.
Determine if it's an epic (Saga output) or plan (Helm output), then run the
appropriate 7-question checklist.
Epic Review (7 Questions):
- Is the goal statement specific enough to test against?
- Are acceptance criteria testable at epic scale?
- Are non-goals explicit?
- Are dependencies identified with owners and dates?
- Are success metrics measurable with existing instrumentation?
- Can an AI agent implement without guessing scope?
- Is the narrative compelling enough to justify the investment?
Plan Review (7 Questions):
- Is the order of work justified?
- Are sprint goals testable (objective + test + constraint)?
- Are dependencies visible and tracked?
- Is estimation reasoning recorded?
- Are unknowns surfaced and buffered for?
- Does the plan reflect learning from previous retros?
- Can the team execute async without daily clarification?
For each item: Pass or Gap with specific feedback.
List all red flags. State what must be fixed before commit.