{
  "$schema_doc": "https://gtmstacker.com/registry/schema/entry.schema.json",
  "stability": "emerging",
  "generator": "agentic-media-registry",
  "generated_at": "2026-09-28T00:00:00Z",
  "id": "com.gtmstacker.registry/tool/hindsight",
  "type": "tool",
  "slug": "hindsight",
  "canonical_url": "https://gtmstacker.com/registry/tool/hindsight/",
  "title": "Hindsight",
  "description": "Hindsight is an agent memory system exposing retain/recall/reflect over four memory networks — world, experience, opinion, observation — with per-bank MCP endpoints and 60+ integrations. Open source: yes (MIT); self-hostable (docker run; PostgreSQL/Oracle AI DB). Free OSS (optional hosted from vectorize.io); ~39.9k stars; active.",
  "category": "ai-infrastructure",
  "tags": [
    "ai-infrastructure",
    "agent-memory",
    "mcp-agents",
    "retrieval",
    "self-hostable",
    "memory"
  ],
  "status": "active",
  "revision": 1,
  "content_hash": "413dc632c7d9ab0119bf6ab006556ce1c8e7b3b0a7e0e3e2a26428df76e090b5",
  "date_published": "2026-09-28T00:00:00Z",
  "date_modified": "2026-09-28T00:00:00Z",
  "source": {
    "name": "vectorize-io · GitHub",
    "url": "https://github.com/vectorize-io/hindsight"
  },
  "license": "MIT",
  "one_liner": "Hindsight gives agents retain/recall/reflect over four memory networks — world, experience, opinion, observation — via per-bank MCP endpoints.",
  "open_source": "yes",
  "self_hostable": "yes",
  "pricing_model": "free",
  "who_its_for": "A team building long-lived agents that need durable, structured memory — separating world knowledge, lived experience, opinions, and observations into distinct banks reached over MCP and self-hosted on a PostgreSQL or Oracle AI Database backend.",
  "who_its_not_for": "A throwaway or single-turn agent that needs no persistent memory, or a team unwilling to run and operate a database-backed memory service.",
  "aliases": [
    "hindsight",
    "vectorize-io/hindsight"
  ],
  "alternatives": [
    "motif",
    "itsuki-memory",
    "cognee"
  ],
  "secondary_categories": [
    "mcp-agents"
  ],
  "last_verified": "2026-09-28",
  "evidence": {
    "claim_type": "vendor-claim",
    "source_id": "https://github.com/vectorize-io/hindsight",
    "note": "MIT per repo; ~39.9k (39,900) stars; ~3,257 commits; active. Exposes retain/recall/reflect over four memory networks (world/experience/opinion/observation) with per-bank MCP endpoints; self-hostable via `docker run` with PostgreSQL or Oracle AI Database backends; 60+ integrations (Claude Code, Cursor, CrewAI, LangGraph, n8n). vectorize.io offers an optional hosted version. Vendor claims 'state-of-the-art on long-term memory tasks' (github.com/vectorize-io/hindsight, verified 2026-09-28)."
  },
  "caveats": "Verified from the primary repo (2026-09-28); no independent testing here. The 'state-of-the-art on long-term memory tasks' claim is the vendor's own, with no benchmark cited here (vendor-claim) — do not treat it as an independent result. Self-hosting requires operating a PostgreSQL or Oracle AI Database backend. vectorize.io offers an optional hosted tier alongside the free OSS.",
  "lead": "Hindsight is an agent memory system exposing retain/recall/reflect over four memory networks — world, experience, opinion, observation — with per-bank MCP endpoints and 60+ integrations. Open source: yes (MIT); self-hostable via `docker run` with PostgreSQL or Oracle AI Database backends; free OSS (vectorize.io offers an optional hosted version). It has…",
  "chunks": [
    {
      "index": 0,
      "heading_path": [],
      "est_tokens": 100,
      "text": "Hindsight is an agent memory system exposing retain/recall/reflect over four memory networks — world, experience, opinion, observation — with per-bank MCP endpoints and 60+ integrations. Open source: yes (MIT); self-hostable via `docker run` with PostgreSQL or Oracle AI Database backends; free OSS (vectorize.io offers an optional hosted version). It has ~39.9k stars and is active (~3,257 commits)."
    },
    {
      "index": 1,
      "heading_path": [
        null,
        "What it does"
      ],
      "est_tokens": 262,
      "text": "retain/recall/reflect over four memory networks — world, experience, opinion, observation — with per-bank MCP endpoints and 60+ integrations. Open source: yes (MIT); self-hostable via `docker run` with PostgreSQL or Oracle AI Database backends; free OSS (vectorize.io offers an optional hosted version). It has ~39.9k stars and is active (~3,257 commits).\n\nHindsight gives an agent a structured, long-term memory rather than a single undifferentiated store. It separates memory into four networks — world (facts about the domain), experience (what happened), opinion (judgments formed), and observation (what was seen) — and exposes three operations across them: retain (write), recall (read), and reflect (synthesize). Each memory bank has its own MCP endpoint, so an agent addresses the right kind of memory directly, and the project ships 60+ integrations including Claude Code, Cursor, CrewAI, LangGraph, and n8n. You self-host it with `docker run` backed by PostgreSQL or an Oracle AI Database. Open source: yes (MIT); self-hostable; free OSS."
    },
    {
      "index": 2,
      "heading_path": [
        null,
        "Provenance"
      ],
      "est_tokens": 257,
      "text": "has its own MCP endpoint, so an agent addresses the right kind of memory directly, and the project ships 60+ integrations including Claude Code, Cursor, CrewAI, LangGraph, and n8n. You self-host it with `docker run` backed by PostgreSQL or an Oracle AI Database. Open source: yes (MIT); self-hostable; free OSS.\n\n- MIT per repo; ~39.9k stars; ~3,257 commits; active. retain/recall/reflect over four memory networks (world/experience/opinion/observation); per-bank MCP endpoints; self-hostable via `docker run` (PostgreSQL / Oracle AI Database); 60+ integrations (Claude Code, Cursor, CrewAI, LangGraph, n8n) (github.com/vectorize-io/hindsight, verified 2026-09-28).\n- vectorize.io offers an optional hosted version; the OSS core is free and self-hostable.\n- Vendor claim: \"state-of-the-art on long-term memory tasks\" — the vendor's own claim (vendor-claim), no benchmark cited here, not independently verified.\n- Curated from the GTM Stacker signal registry (2026-09-28 pass); license/facts independently verified 2026-09-28."
    },
    {
      "index": 3,
      "heading_path": [
        null,
        "Why it matters for a GTM stack"
      ],
      "est_tokens": 304,
      "text": "2026-09-28). - vectorize.io offers an optional hosted version; the OSS core is free and self-hostable. - Vendor claim: \"state-of-the-art on long-term memory tasks\" — the vendor's own claim (vendor-claim), no benchmark cited here, not independently verified. - Curated from the GTM Stacker signal registry (2026-09-28 pass); license/facts independently verified 2026-09-28.\n\nA GTM agent that forgets is a liability: it re-asks answered questions, loses account context between sessions, and cannot build a durable view of a prospect. Hindsight offers a shaped memory for exactly that — separating world facts, lived experience, formed opinions, and raw observations, each addressable over MCP — so an outbound or account agent can retain, recall, and reflect across a long relationship, self-hosted on your own PostgreSQL. Open source: yes (MIT); self-hostable; free OSS. The honest read: the four-network model plus per-bank MCP is a serious, well-adopted design (~39.9k stars, 60+ integrations), but \"state-of-the-art on long-term memory\" is the vendor's unbenchmarked claim, and self-hosting means running a real database — evaluate recall quality on your own data before it becomes the agent's source of truth."
    }
  ],
  "alternates": {
    "markdown": "https://gtmstacker.com/registry/tool/hindsight/index.md",
    "html": "https://gtmstacker.com/registry/tool/hindsight/",
    "json": "https://gtmstacker.com/registry/tool/hindsight/index.json",
    "server_json": "https://gtmstacker.com/registry/tool/hindsight/server.json"
  },
  "jsonld": {
    "@context": "https://schema.org",
    "@graph": [
      {
        "@type": "WebSite",
        "@id": "https://gtmstacker.com/#website",
        "url": "https://gtmstacker.com/",
        "name": "GTM Stacker Agent Registry",
        "description": "A daily-updated, agent-native registry of open-source tool discoveries, tool updates, and curated news for the go-to-market / RevOps engineering niche. Machine-readable first: agents can discover, parse, page, and delta-sync it without scraping HTML.",
        "inLanguage": "en",
        "publisher": {
          "@id": "https://gtmstacker.com/#organization"
        }
      },
      {
        "@type": "Organization",
        "@id": "https://gtmstacker.com/#organization",
        "name": "GTM Stacker",
        "url": "https://gtmstacker.com",
        "description": "The growth-systems practice of Theo Popov: AI-native enrichment, outbound, content engines and internal tooling for startups and venture programs. Its agent-native media property, the GTM Stacker Agent Registry, maintains a daily-updated catalog of open-source go-to-market and RevOps tools that both people and AI engines can discover, compare, and cite.",
        "foundingDate": "2024-08",
        "knowsAbout": [
          "go-to-market engineering",
          "RevOps",
          "sales automation",
          "marketing operations",
          "open-source software",
          "AI agents"
        ],
        "founder": {
          "@type": "Person",
          "@id": "https://gtmstacker.com/#founder",
          "name": "Theo Popov",
          "jobTitle": "Growth Operations & GTM Systems",
          "url": "https://gtmstacker.com/about/",
          "sameAs": [
            "https://www.linkedin.com/in/theo-popov",
            "https://x.com/Theo_Popov",
            "https://github.com/theopopov"
          ],
          "worksFor": {
            "@id": "https://gtmstacker.com/#organization"
          }
        },
        "sameAs": [
          "https://www.linkedin.com/company/gtmstacker",
          "https://www.youtube.com/@gtmstacker",
          "https://www.instagram.com/gtmstacker/",
          "https://www.tiktok.com/@gtmstacker"
        ],
        "mainEntityOfPage": "https://gtmstacker.com/registry/about/"
      },
      {
        "@type": "SoftwareApplication",
        "@id": "https://gtmstacker.com/registry/tool/hindsight/#software",
        "name": "Hindsight",
        "identifier": "vectorize-io/hindsight",
        "description": "Hindsight is an agent memory system exposing retain/recall/reflect over four memory networks — world, experience, opinion, observation — with per-bank MCP endpoints and 60+ integrations. Open source: yes (MIT); self-hostable (docker run; PostgreSQL/Oracle AI DB). Free OSS (optional hosted from vectorize.io); ~39.9k stars; active.",
        "applicationCategory": "DeveloperApplication",
        "url": "https://gtmstacker.com/registry/tool/hindsight/",
        "datePublished": "2026-09-28T00:00:00Z",
        "dateModified": "2026-09-28T00:00:00Z",
        "isPartOf": {
          "@id": "https://gtmstacker.com/#website"
        },
        "license": "https://spdx.org/licenses/MIT.html",
        "codeRepository": "https://github.com/vectorize-io/hindsight",
        "keywords": "ai-infrastructure, mcp-agents, agent-memory, retrieval, self-hostable, memory",
        "author": {
          "@type": "Organization",
          "name": "vectorize-io",
          "url": "https://github.com/vectorize-io",
          "sameAs": [
            "https://github.com/vectorize-io/hindsight"
          ]
        },
        "offers": {
          "@type": "Offer",
          "price": 0,
          "priceCurrency": "USD"
        },
        "isSimilarTo": [
          {
            "@type": "SoftwareApplication",
            "name": "motif",
            "url": "https://gtmstacker.com/registry/tool/motif/",
            "applicationCategory": "DeveloperApplication",
            "offers": {
              "@type": "Offer",
              "price": 0,
              "priceCurrency": "USD"
            }
          },
          {
            "@type": "SoftwareApplication",
            "name": "Itsuki",
            "url": "https://gtmstacker.com/registry/tool/itsuki-memory/",
            "applicationCategory": "DeveloperApplication",
            "offers": {
              "@type": "Offer",
              "price": 0,
              "priceCurrency": "USD"
            }
          },
          {
            "@type": "SoftwareApplication",
            "name": "Cognee",
            "url": "https://gtmstacker.com/registry/tool/cognee/",
            "applicationCategory": "DeveloperApplication",
            "offers": {
              "@type": "Offer",
              "price": 0,
              "priceCurrency": "USD"
            }
          }
        ]
      },
      {
        "@type": "BreadcrumbList",
        "@id": "https://gtmstacker.com/registry/tool/hindsight/#breadcrumb",
        "itemListElement": [
          {
            "@type": "ListItem",
            "position": 1,
            "name": "GTM Stacker Registry",
            "item": "https://gtmstacker.com/registry/"
          },
          {
            "@type": "ListItem",
            "position": 2,
            "name": "AI Infrastructure",
            "item": "https://gtmstacker.com/registry/category/ai-infrastructure/"
          },
          {
            "@type": "ListItem",
            "position": 3,
            "name": "Hindsight",
            "item": "https://gtmstacker.com/registry/tool/hindsight/"
          }
        ]
      }
    ]
  },
  "tool": {
    "name": "vectorize-io/hindsight",
    "repository": {
      "url": "https://github.com/vectorize-io/hindsight",
      "source": "github"
    }
  }
}