What data we collect
MyManual only stores data you explicitly provide during the guided interview. We collect:
- Your Slack user ID (to associate your manual with your account)
- Your workspace/team ID
- The content of your manual (communication preferences, feedback style, focus hours, misreads, energizers/drainers, ask-me-about items)
- Per-field provenance (whether each entry was self-authored or suggested from your messages and confirmed by you)
- Publication status and read policy settings
What we do NOT collect
- We do not collect medical information, diagnoses, or protected characteristics
- We do not store your Slack messages — Real-Time Search results are processed in-memory and discarded
- We do not collect data about other users without their explicit participation
- We do not use any data for LLM training
Real-Time Search (RTS)
The optional self-draft feature uses Slack's Real-Time Search API to analyze your own public messages only, only when you explicitly request it. Search results are processed in-memory to generate suggestions and are never stored. Every suggestion requires your explicit confirmation before entering your manual.
How data is stored
- Your manual data is stored in a Cloudflare D1 database (SQLite)
- Your manual is also stored as a Slack Canvas that you own
- Data is encrypted in transit (HTTPS) and at rest (D1 encryption)
Who can see your manual
- Private (default): Only you
- Published: Anyone in your Slack workspace who runs
/manual @you or queries the MCP server
- You control your publication status at all times
- You can optionally be notified when someone reads your manual
MCP server access
When your manual is published, it is accessible via our MCP server to other AI agents in your workspace. Only published manuals are exposed. The MCP server provides read-only access — no agent can modify your manual.
Data deletion
You can delete your manual at any time. Deletion removes all data from our database and the associated Slack Canvas. This action is irreversible.
Third-party services
- Slack: We use Slack's APIs to power the agent, store canvases, and search messages. Slack's privacy policy applies to data within Slack.
- OpenRouter: We use OpenRouter to route LLM requests for the interview ghostwriter and Q&A features. Your manual content is processed by the configured LLM to generate polished text. We do not store LLM inputs or outputs beyond your confirmed manual entries.
- Cloudflare: Our infrastructure runs on Cloudflare Workers and D1. Cloudflare's privacy policy applies to infrastructure-level data.
Contact
For privacy questions, contact us at support@mymanual.app.