Skip to content
Beta Multi-model support (OpenAI, Gemini, and more) launching soon.
RBA RBA Skills

/personalization-plan

Installable Ships in personalization-planner — or copy the raw command below.
Open Claude.ai

Paste into Claude Code, Claude.ai, the desktop app, or the browser extension — ⌘/Ctrl+V.

ExperimentalNo security review

Route the request based on the argument provided:

Routing logic

No argument or full: Run the full pipeline via personalization-planner-agent. Collect intake interactively, then produce hypothesis backlog + audience definition + test plan + measurement spec + assembled experiment brief.

hypothesis <brief-or-file>: Run hypothesis-generator only. Produce a signal inventory, problem map, and prioritized hypothesis backlog. Output: hypothesis-backlog-<slug>-<YYYY-MM>.md.

audience <persona-or-segment-description>: Run audience-segment-builder only. Define behavioral/firmographic/lifecycle segments for the named audience and produce platform-specific configuration. Output: audience-segments-<slug>-<YYYY-MM>.md.

test-plan <HYP-NNN-or-hypothesis-brief>: Run experiment-test-plan only. Calculate sample size, define traffic split, write variant specs, produce QA checklist and stopping rules. Output: test-plan-<hypothesis-id>-<YYYY-MM>.md.

measure <HYP-NNN-or-test-plan-path>: Run personalization-measurement-spec only. Author primary metric, secondary metrics, guardrail metrics, counter-metric analysis, data readiness checklist, and reporting spec. Output: measurement-spec-<hypothesis-id>-<YYYY-MM>.md.

review <client-context-or-'interview'>: Run experimentation-program-review. Assess the client's experimentation program maturity and produce a scored assessment + improvement roadmap. Output: experimentation-program-review-<client>-<YYYY-MM>.md.

Context reading

Before routing, check for any of these files in the current directory and read them automatically if found:

Glob: *.analytics.json, *-brief.md, *-research.md, *journey*.md, *persona*.md, hypothesis-backlog-*.md, test-plan-*.md

If context files are found, pass their contents to the routed skill as input so it does not start from scratch.

Output

After routing and execution, always summarize:

  1. Which skill(s) were run
  2. Which files were created (with full paths)
  3. Any assumptions made that the user should review
  4. Recommended next step (which skill to run next, or "ready for platform setup")
Technical details
Source
personalization-planner
Author
Dakota Smith
Invocation
/personalization-plan
Maturity
experimental
Security review
No security review
Owners
  • strategy
  • engineering
File
plugins/personalization-planner/commands/personalization-plan.md

Edit this command on GitHub — opens GitHub's web editor. No git knowledge needed.