Your archive has an interface
A CLI, a REST API, and an MCP server. Your saved reading is not stuck behind someone else’s UI — query it, pipe it, hand it to an agent, build on top of it.
A read-later archive with a CLI, a REST API, and an MCP server — so everything you’ve saved is one query away, for you or your agent.
# save an article and keep a copy of the source
marca add https://example.com/essay --article --archive
# search everything you have ever saved
marca search "rate limiting" --limit 5
# it speaks JSON, so it composes
marca list --tag reading --json | jq '.[].url' Save, search, and pipe your library without leaving the terminal.
A CLI, a REST API, and an MCP server. Your saved reading is not stuck behind someone else’s UI — query it, pipe it, hand it to an agent, build on top of it.
Marca keeps the source when you save it, not just the link. Annotate what matters. Find it later by phrase, tag, title, or the tiny detail you can almost remember.
Export everything whenever you want. No intrusive tracking. AI features stay off until you turn them on.
Then $3 a month, or $29 a year if you’d rather not think about it again. Cancel whenever — your export goes with you.