Skip to content
Agent Memory
Command

/prd-split

Split the approved PRD into one issue spec per manifest entry

BOOST
From plugin
kipi-system
12823 skills5 agents23 commands5 MCP
Install
> /plugin marketplace add assafkip/kipi-system

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/prd-split

Context preview

What this command does when you run it.

Split the approved PRD into one issue spec per manifest entry

Command definition

prd-split.md
description: Split the approved PRD into one issue spec per manifest entry
allowed-tools: Bash

Materialize the PRD's `## Issues` fenced JSON manifest into issue specs under `issues_dir`. The split script enforces:

  • PRD status is `approved`.
  • Every entry has non-empty `allowed_files` and non-empty `required_checks`.

(An empty `required_checks` silently bypasses the runner's verification gate — the split refuses to create such specs.)

  • No existing issue file is clobbered. Re-runs are idempotent only when

byte-identical.

Two-step flow — dry-run first so the author can see planned writes before any file lands on disk:

1. Dry-run:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/prd_split.py" --dry-run

Read the `would_create` paths back to the author. Confirm before committing. If they do not confirm, stop. Do NOT run step 2 without explicit agreement.

2. Commit:

python3 "${CLAUDE_PLUGIN_ROOT}/scripts/prd_split.py"

On collision, the script names the divergent files. Do not delete or overwrite them. Ask the author to rename either the manifest entry or the existing file.

After a successful split, tell the author to archive or clear the PRD before starting an issue (`prd_runner.py archive` or `prd_runner.py clear`). The concurrency guard blocks loading an issue while a PRD context is active.

Ships withkipi-system

Give your AI the right knowledge, at the moment it needs it. Kipi turns the scattered knowledge around your work into context that follows you across AI sessions: decisions, conversations, projects, people, commitments, lessons, documents, history.

Get the whole plugin

Other commands on kipi-system.