Disclaimer

At its core, librocat (a Laughing Hermit, Inc. product) is a library and information science tool: it catalogs, indexes, and serves the knowledge you and your agents put into it. It applies classification, controlled vocabulary, and weeding practice to Markdown files. It is infrastructure, not an oracle.

librocat stores and retrieves what you give it. It does not verify, generate, or endorse the content of your concepts, and it does not make decisions. Your agents and your team do. You are responsible for the accuracy of the knowledge you store, for the actions your agents take with it, and for any decision made on top of it.

In the hosted library, Libro, the library cat, fills in a description and tags your agent left blank, picking a sentence from the concept and tags from your own vocabulary, and once a week proposes supersessions, retirements, merges, synonyms, review dates, and gaps on the review shelf. A catalog record is not a judgment, and it can be wrong. Nothing on the review shelf changes until you or your agent approves it, and every write is a revision, so you can always see and undo what changed.

Nothing stored in or retrieved from librocat is legal, medical, financial, or engineering advice. It is what you and your agents put there.

The service and the open-source software are provided "as is", without warranty of any kind, express or implied, including fitness for a particular purpose. The software, self-hosted or not, is licensed under Apache-2.0, and its license text governs.

To the maximum extent permitted by law, Laughing Hermit, Inc. is not liable for indirect, incidental, or consequential damages, or for losses arising from decisions made by you, your team, or your AI agents using knowledge stored in or retrieved from librocat.

librocat does not currently hold SOC 2 or ISO 27001 certification. If your organization needs formal attestations, contact us about the Enterprise plan before storing regulated data.