Integrations
Spectr reads the tools your project already uses for context, and publishes review-ready stories back into the tracker your team works in. Most connections are scoped to a single project; your AI agents and coding tools connect to you, under User Settings > Connections. Every credential is encrypted at rest.
Where the conversation comes from. Import a call and Spectr reads its notes, action items, and transcript.
Import a Fathom call and get structured specs and stories from its notes, action items, and transcript
Native Granola import - no Zapier, no scripts - straight through to Jira, Linear or Shortcut
Connect Read.ai over OAuth and mix its meetings freely with Fathom and Granola in one project
Import a Fireflies meeting and get structured specs and stories from its recap, action items, and transcript
Where your stories land when you publish them.
What Spectr reads before it writes, so a story references your real docs, designs, and code.
Pin up to ten Notion pages and Spectr reads them before it writes a story
Pin up to five Figma files and Spectr writes stories against the screens themselves
Pin up to five repos so Spectr writes stories that match the code that already exists
Pin Google Docs, Sheets and Slides through the Google Picker as spec context
Read-only budget, schedule, and logged hours next to the delivery work they pay for.
Read-only errors, feedback, and rollout state — cited with a date in the specs Spectr writes.
Start a coding agent on your project's stories and tasks from a Development workstream, on your own account, and track its run on the board.
Your own AI agents, reading and acting on Spectr over MCP as you. Connected under your own account, not a project.
Connect Claude over MCP to read - and, with your permission, act on - your Spectr data as you
Connect OpenAI Codex over MCP to read - and, with your permission, act on - your Spectr data as you
Connect Cursor over MCP to read - and, with your permission, act on - your Spectr data as you
Connect ChatGPT over MCP to read your Spectr data as you
End to end
A call is imported from a note-taker (Fathom, Granola, Read.ai, or Fireflies.ai). Spectr picks out a requirement, a bug, an action item and an out-of-scope ask, reads Notion, Figma, and GitHub for context, and drafts a story and a bug with acceptance criteria. A person approves both, they are published to the team's tracker (Jira, Linear, or Shortcut), and the requirements, scope log and project memory update from the same call.
The end-to-end route, one page per combination — what gets read, what gets written, and what is specific to that pairing.
