Product managers: connect PRDs, research, and roadmap so the agent can actually reason
User interviews, competitive analyses, requirements, PRDs, roadmap — all in the Kition vault. The agent reviewing a requirement can pull up the interview from three months ago, the last comp scan, and the current spec — and give a defensible take.
Stop asking AI to "help me draft a PRD." Ask it to "give me a recommendation based on all my research."
Who it is for
B2B / B2C PMs, indie makers, designer-PM hybrids.
What hurts today
Requirements in Linear, interviews in Notion, PRDs in Google Docs, roadmap in Figma. The AI only sees what you copy into its prompt.
How Kition helps
Kition consolidates everything into one vault. The agent cites a specific interview quote, comp page, or prior decision. Linear, Jira, Figma plug in via MCP — the agent can read those SaaS sources directly.
Suggested workflow
Step 1
Interviews → Markdown
One .md per interview, tagged with persona, key quotes, and action items.
Step 2
Backlog table
Track requirements in a table; the agent prioritizes by frequency in interviews.
Step 3
PRDs
When the agent drafts a PRD it cites real interviews, competitor scans, and existing data.
Step 4
MCP for Linear
Sync states and bulk-create issues from the agent.
Outcomes
- Requirement provenance is traceable end-to-end
- PRDs cite research rather than vibes
- AI recommendations have evidence
- Five tools become one
Other roles
Ready when you are.
Kition is a local-first AI workspace. Markdown documents, structured tables, and an AI agent — running on your own machine, against the model provider you choose.