{"id":"hindsight","name":"Hindsight","maker":"Vectorize","url":"https://hindsight.vectorize.io","repo":"vectorize-io/hindsight","docs":"https://hindsight.vectorize.io","summary":"An agent memory system that stores facts, experiences and learned beliefs, with retrieval and reflection operations.","details":"Hindsight runs as a server, usually in Docker, with Python and TypeScript clients, a command-line tool and an MCP interface. It separates raw facts from observations the agent forms over time. The same code is available as a hosted offering from its maker.","open_source":true,"hosted":true,"speaks":["mcp"],"verified":"2026-11-15","category":"memory","category_name":"Agent memory","page":"https://modelmoot.com/tools/hindsight","reviewed":"2026-11-15","checked":{"at":"2026-11-15T14:03:44.997Z","address_answers":true,"repository":{"at":"2026-11-16T00:27:04.284Z","archived":false,"licence":"MIT","language":"Python","last_change":"2026-11-14T20:17:16.000Z","latest_release":{"tag":"v1.2.1","published":"2026-11-10T14:34:26.000Z"}}},"note":"Descriptions are the site's own. Facts under \"checked\" were read automatically from the tool's address and public repository."}