v2026.10.11

PM-Kit now helps choose the smallest useful method for a product decision, with stronger evidence, risk, and outcome guidance.

Added

• /prioritize: direct triage, MoSCoW, RICE, or Kano selected by the decision and available evidence, with strategic checks before scoring.
• /discover: root-cause investigation, Input–Process–Output, JTBD, research guidance, opportunity mapping, and investment-risk tests.
• /journey: evidence-linked user journey maps that distinguish observed, inferred, and proposed paths.
• /show-me: focused visual explanations, with explicit routing for Mermaid and Excalidraw requests.
• Core judgment and knowledge guides covering framework selection, source integrity, assumptions, product risks, outcome measures, and human-reviewed lessons.
• Amp orb setup and resume scripts with locked document-tool dependencies and reuse of warm installations.

Changed

• /explain adapts to the user's language and requested depth; explanations stay in chat unless saving is requested.
• /doc connects problems and evidence to assumptions, user stories, validation, and relevant outcome measures without forcing every document into a PRD.
• /weekly separates delivery, learning, and measured outcomes, then proposes priorities and evidence-backed decision changes.
• Both framework updaters distribute the new core guides while preserving personal identity, configuration, and user notes.
• README, onboarding, workflow guidance, and help documentation describe the new workflows and approval boundaries.

Known Limitations

• Existing YAML frontmatter issues in /daily, /block, and /decide remain unchanged and may prevent skill discovery in hosts that enforce YAML parsing.
• Framework selection is agent guidance, not a deterministic runtime or a guarantee of consistent decisions.
• Claude Code marketplace installation, a custom UI/runtime, and a RAG platform are not included in this release.