By Max Techera · Open-source (MIT) · Updated June 2026
Incident postmortems

Share an incident postmortem with customers privately

Give affected customers the full timeline without publishing your internals to the world.

npxnpx hushdrop report.html

After an outage, your biggest customers want the full story: timeline, root cause, what changes. That level of detail doesn't belong on a public status page, but it shouldn't be a messy email thread either.

A private postmortem page lets you include architecture details and graphs that you'd never put on a public blog, and you control exactly who can read it.

Share it in one step

hush postmortem-2026-11-18.html -s inc-1118 --expire 60d

Your agent can draft it from the incident channel and logs, then publish it locked when you say "share the postmortem with the affected accounts".

Settings that fit

FAQ

Can I update the postmortem as we learn more?
Yes. Republish with the same slug and password (-s inc-1118 -p ...) and the same link serves the update.
Can my AI agent do this for me?
Yes. Hushdrop ships as a CLI, an MCP server and a SKILL.md, so Claude Code, Codex, Cursor, Gemini CLI or anything that can run a shell command or call an MCP tool can publish for you. Tell it "share this with the client, locked" and it returns the link and password.

Turn a page into a link right now

Free, no account. Paste HTML here, or connect your agent once and it publishes for you.

Paste HTML, get a link

Give your agent a publish button

Claude Code:

claude mcp add --transport http hushdrop https://hushdrop.dev/mcp

claude.ai, ChatGPT, Cursor and others: add https://hushdrop.dev/mcp as a remote MCP server, then ask your agent to "share this".