Plan Review Mode

Read workflow.md for detailed phases and anti-patterns.

Conduct a John Carmack-level review of epic plans.

Role: Code Review Coordinator (NOT the reviewer) Backends: RepoPrompt (rp) or Codex CLI (codex)

CRITICAL: flowctl is BUNDLED — NOT installed globally. which flowctl will fail (expected). Always use:

FLOWCTL="${DROID_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT}}/scripts/flowctl"

Backend Selection

Priority (first match wins):

• --review=rp|codex|export|none argument

• FLOW_REVIEW_BACKEND env var (rp , codex , none )

• .flow/config.json → review.backend

• Error - no auto-detection

Parse from arguments first

Check $ARGUMENTS for:

• --review=rp or --review rp → use rp

• --review=codex or --review codex → use codex

• --review=export or --review export → use export

• --review=none or --review none → skip review

If found, use that backend and skip all other detection.

Otherwise read from config

Priority: --review flag > env > config

BACKEND=$($FLOWCTL review-backend)

if [[ "$BACKEND" == "ASK" ]]; then echo "Error: No review backend configured." echo "Run /flow-next:setup to configure, or pass --review=rp|codex|none" exit 1 fi

echo "Review backend: $BACKEND (override: --review=rp|codex|none)"

Critical Rules

For rp backend:

• DO NOT REVIEW THE PLAN YOURSELF - you coordinate, RepoPrompt reviews

• MUST WAIT for actual RP response - never simulate/skip the review

• MUST use setup-review

• handles window selection + builder atomically

• DO NOT add --json flag to chat-send - it suppresses the review response

• Re-reviews MUST stay in SAME chat - omit --new-chat after first review

For codex backend:

• Use $FLOWCTL codex plan-review exclusively

• Pass --receipt for session continuity on re-reviews

• Parse verdict from command output

For all backends:

• If REVIEW_RECEIPT_PATH set: write receipt after review (any verdict)

• Any failure → output <promise>RETRY</promise> and stop

FORBIDDEN:

• Self-declaring SHIP without actual backend verdict

• Mixing backends mid-review (stick to one)

• Skipping review when backend is "none" without user consent

Input

Arguments: $ARGUMENTS Format: <flow-epic-id> [focus areas]

Workflow

See workflow.md for full details on each backend.

FLOWCTL="${DROID_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT}}/scripts/flowctl" REPO_ROOT="$(git rev-parse --show-toplevel 2>/dev/null || pwd)"

Step 0: Detect Backend

Run backend detection from SKILL.md above. Then branch:

Codex Backend

EPIC_ID="${1:-}" RECEIPT_PATH="${REVIEW_RECEIPT_PATH:-/tmp/plan-review-receipt.json}"

Save checkpoint before review (recovery point if context compacts)

$FLOWCTL checkpoint save --epic "$EPIC_ID" --json

--files: comma-separated CODE files for reviewer context

Epic/task specs are auto-included; pass files the plan will CREATE or MODIFY

How to identify: read the epic spec, find files mentioned or directories affected

Example: epic touches auth → pass existing auth files for context

Dynamic approach (if epic mentions specific paths):

CODE_FILES=$(grep -oE 'src/[^ ]+.(ts|py|js)' .flow/specs/${EPIC_ID}.md | sort -u | paste -sd,)

Or list key files manually:

CODE_FILES="src/main.py,src/config.py"

$FLOWCTL codex plan-review "$EPIC_ID" --files "$CODE_FILES" --receipt "$RECEIPT_PATH"

Output includes VERDICT=SHIP|NEEDS_WORK|MAJOR_RETHINK

On NEEDS_WORK: fix plan via $FLOWCTL epic set-plan AND sync affected task specs via $FLOWCTL task set-spec , then re-run (receipt enables session continuity).

Note: codex plan-review automatically includes task specs in the review prompt.

RepoPrompt Backend

⚠️ STOP: You MUST read and execute workflow.md now.

Go to the "RepoPrompt Backend Workflow" section in workflow.md and execute those steps. Do not proceed here until workflow.md phases are complete.

The workflow covers:

• Get plan content and save checkpoint

• Atomic setup (setup-review) → sets $W and $T

• Augment selection (epic + task specs)

• Send review and parse verdict

Return here only after workflow.md execution is complete.

Fix Loop (INTERNAL - do not exit to Ralph)

CRITICAL: Do NOT ask user for confirmation. Automatically fix ALL valid issues and re-review — our goal is production-grade world-class software and architecture. Never use AskUserQuestion in this loop.

If verdict is NEEDS_WORK, loop internally until SHIP:

• Parse issues from reviewer feedback

• Fix epic spec (stdin preferred, temp file if content has single quotes):

Preferred: stdin heredoc

$FLOWCTL epic set-plan <EPIC_ID> --file - --json <<'EOF' <updated epic spec content> EOF

Or temp file

$FLOWCTL epic set-plan <EPIC_ID> --file /tmp/updated-plan.md --json

• Sync affected task specs - If epic changes affect task specs, update them: $FLOWCTL task set-spec <TASK_ID> --file - --json <<'EOF' <updated task spec content> EOF

Task specs need updating when epic changes affect:

• State/enum values referenced in tasks

• Acceptance criteria that tasks implement

• Approach/design decisions tasks depend on

• Lock/retry/error handling semantics

• API signatures or type definitions

• Re-review:

• Codex: Re-run flowctl codex plan-review (receipt enables context)

• RP: $FLOWCTL rp chat-send --window "$W" --tab "$T" --message-file /tmp/re-review.md (NO --new-chat )

• Repeat until <verdict>SHIP</verdict>

Recovery: If context compaction occurred during review, restore from checkpoint:

$FLOWCTL checkpoint restore --epic <EPIC_ID> --json

CRITICAL: For RP, re-reviews must stay in the SAME chat so reviewer has context. Only use --new-chat on the FIRST review.