GHSA-47PJ-3JCM-6WHG

Vulnerability from github – Published: 2026-08-06 19:03 – Updated: 2026-08-06 19:03
VLAI
Summary
LangGraph: Namespace prefix matching crosses segment boundaries in Postgres and SQLite stores
Details

Summary

The Postgres and SQLite stores persist hierarchical namespaces as a dot-joined string (("memories", "alice") becomes memories.alice) and scoped reads by matching that string with LIKE '<path>%'. Because LIKE has no notion of the . separator, a scoped search or list_namespaces also matched sibling namespaces whose flattened form shares leading characters.

Applications commonly use the namespace as a tenant boundary. Where they do, a read scoped to one namespace could return items belonging to another, without any crafted input — an ordinary scoped request was sufficient.

We have no evidence of this behavior being exploited in the wild.

Affected users / systems

You may be affected if you:

  • use PostgresStore/AsyncPostgresStore or SqliteStore/AsyncSqliteStore, and
  • rely on the namespace to separate data between users or tenants, and
  • have namespace labels where one is a prefix of another (1 and 12, alice and alice2), or labels containing _ or %

Applications whose namespace labels are fixed-length identifiers such as UUIDs, containing no _ or %, are not affected — no such label can be a prefix of another. InMemoryStore compares namespaces element-wise and is not affected.

Three distinct cases were possible:

  • Sibling namespaces. A read scoped to ("foo",) also returned items under ("foobar",) and ("foo2",).
  • Unescaped pattern metacharacters. _ and % are legal namespace labels — only . is rejected — but were interpolated into the match pattern unescaped, so ("user_1",) also matched ("userX1",).
  • Suffix conditions. list_namespaces(suffix=("alice",)) also matched the sibling leaf users.malice.

This is not SQL injection. Values were passed as bound parameters and never interpolated into statement text; the bound value was itself a LIKE pattern whose metacharacters were not neutralized.

Impact

  • Confidentiality: disclosure of stored items belonging to namespaces outside the caller's intended scope, where namespaces are used as a tenant or user boundary.
  • No integrity or availability impact. get, put, and delete compare namespaces with = and were never affected; the issue is limited to read paths.

Patches / mitigation

Prefix scoping now matches the namespace exactly or requires the . separator before any remainder, pattern metacharacters in labels are escaped, and list_namespaces uses segment-aware matching for both prefix and suffix conditions.

On SQLite, the descendant match moved from LIKE to GLOB. LIKE is case-insensitive for ASCII in SQLite, so scoped reads previously matched namespaces differing only in case, while get/put/delete treated them as distinct. Search now agrees with them.

Upgrade to langgraph-checkpoint-postgres 3.1.1 or langgraph-checkpoint-sqlite 3.1.1.

Compatibility

* in a list_namespaces match path now spans exactly one namespace segment. This restores the documented behavior — NamespacePath documents ("cache", "*", "v1") as "any cache category with v1 version" — and matches InMemoryStore. Multi-segment matching was an artifact of translating * into a SQL % wildcard, the same mechanism responsible for this issue, and could not be preserved while fixing it.

Callers relying on the previous behavior can express "match at any depth" by combining both match conditions, which are ANDed:

list_namespaces(prefix=["uid"], suffix=["alice"])

Applications whose namespace labels cannot be prefixes of one another see no behavioral change.

Operational guidance

  • Prefer fixed-length namespace labels such as UUIDs, so no label can be a prefix of another.
  • Where labels are user-supplied, validate them at the boundary rather than relying on scoping alone.

LangSmith / hosted deployments note

Unlike previous store advisories, this issue does reach hosted deployments. LangSmith deployments default to LANGGRAPH_STORE_BACKEND=python, which uses AsyncPostgresStore from checkpoint-postgres. Deployments configured with LANGGRAPH_STORE_BACKEND=grpc use a separate implementation that received an equivalent fix.

Severity
5.3 (Medium)
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:N/A:N
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "langgraph-checkpoint-postgres"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "3.1.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "langgraph-checkpoint-sqlite"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "3.1.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-71433"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-200",
      "CWE-863"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-08-06T19:03:12Z",
    "nvd_published_at": null,
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nThe Postgres and SQLite stores persist hierarchical namespaces as a dot-joined string (`(\"memories\", \"alice\")` becomes `memories.alice`) and scoped reads by matching that string with `LIKE \u0027\u003cpath\u003e%\u0027`. Because `LIKE` has no notion of the `.` separator, a scoped `search` or `list_namespaces` also matched sibling namespaces whose flattened form shares leading characters.\n\nApplications commonly use the namespace as a tenant boundary. Where they do, a read scoped to one namespace could return items belonging to another, without any crafted input \u2014 an ordinary scoped request was sufficient.\n\nWe have no evidence of this behavior being exploited in the wild.\n\n## Affected users / systems\n\nYou may be affected if you:\n\n- use `PostgresStore`/`AsyncPostgresStore` or `SqliteStore`/`AsyncSqliteStore`, and\n- rely on the namespace to separate data between users or tenants, and\n- have namespace labels where one is a prefix of another (`1` and `12`, `alice` and `alice2`), or labels containing `_` or `%`\n\nApplications whose namespace labels are fixed-length identifiers such as UUIDs, containing no `_` or `%`, are not affected \u2014 no such label can be a prefix of another. `InMemoryStore` compares namespaces element-wise and is not affected.\n\nThree distinct cases were possible:\n\n- **Sibling namespaces.** A read scoped to `(\"foo\",)` also returned items under `(\"foobar\",)` and `(\"foo2\",)`.\n- **Unescaped pattern metacharacters.** `_` and `%` are legal namespace labels \u2014 only `.` is rejected \u2014 but were interpolated into the match pattern unescaped, so `(\"user_1\",)` also matched `(\"userX1\",)`.\n- **Suffix conditions.** `list_namespaces(suffix=(\"alice\",))` also matched the sibling leaf `users.malice`.\n\nThis is not SQL injection. Values were passed as bound parameters and never interpolated into statement text; the bound value *was itself* a `LIKE` pattern whose metacharacters were not neutralized.\n\n## Impact\n\n- Confidentiality: disclosure of stored items belonging to namespaces outside the caller\u0027s intended scope, where namespaces are used as a tenant or user boundary.\n- No integrity or availability impact. `get`, `put`, and `delete` compare namespaces with `=` and were never affected; the issue is limited to read paths.\n\n## Patches / mitigation\n\nPrefix scoping now matches the namespace exactly or requires the `.` separator before any remainder, pattern metacharacters in labels are escaped, and `list_namespaces` uses segment-aware matching for both prefix and suffix conditions.\n\nOn SQLite, the descendant match moved from `LIKE` to `GLOB`. `LIKE` is case-insensitive for ASCII in SQLite, so scoped reads previously matched namespaces differing only in case, while `get`/`put`/`delete` treated them as distinct. Search now agrees with them.\n\nUpgrade to `langgraph-checkpoint-postgres` 3.1.1 or `langgraph-checkpoint-sqlite` 3.1.1.\n\n## Compatibility\n\n`*` in a `list_namespaces` match path now spans exactly one namespace segment. This restores the documented behavior \u2014 `NamespacePath` documents `(\"cache\", \"*\", \"v1\")` as \"any cache category with v1 version\" \u2014 and matches `InMemoryStore`. Multi-segment matching was an artifact of translating `*` into a SQL `%` wildcard, the same mechanism responsible for this issue, and could not be preserved while fixing it.\n\nCallers relying on the previous behavior can express \"match at any depth\" by combining both match conditions, which are ANDed:\n\n```python\nlist_namespaces(prefix=[\"uid\"], suffix=[\"alice\"])\n```\n\nApplications whose namespace labels cannot be prefixes of one another see no behavioral change.\n\n## Operational guidance\n\n- Prefer fixed-length namespace labels such as UUIDs, so no label can be a prefix of another.\n- Where labels are user-supplied, validate them at the boundary rather than relying on scoping alone.\n\n## LangSmith / hosted deployments note\n\nUnlike previous store advisories, this issue does reach hosted deployments. LangSmith deployments default to `LANGGRAPH_STORE_BACKEND=python`, which uses `AsyncPostgresStore` from `checkpoint-postgres`. Deployments configured with `LANGGRAPH_STORE_BACKEND=grpc` use a separate implementation that received an equivalent fix.",
  "id": "GHSA-47pj-3jcm-6whg",
  "modified": "2026-08-06T19:03:12Z",
  "published": "2026-08-06T19:03:12Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/langchain-ai/langgraph/security/advisories/GHSA-47pj-3jcm-6whg"
    },
    {
      "type": "WEB",
      "url": "https://github.com/langchain-ai/langgraph/pull/8478"
    },
    {
      "type": "WEB",
      "url": "https://github.com/langchain-ai/langgraph/commit/66ebe1a0da921e73f0f9f879ba105d314c079f7c"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/langchain-ai/langgraph"
    },
    {
      "type": "WEB",
      "url": "https://github.com/langchain-ai/langgraph/releases/tag/checkpointpostgres%3D%3D3.1.1"
    },
    {
      "type": "WEB",
      "url": "https://github.com/langchain-ai/langgraph/releases/tag/checkpointsqlite%3D%3D3.1.1"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "LangGraph: Namespace prefix matching crosses segment boundaries in Postgres and SQLite stores"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.





Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.


Detection rules are retrieved from Rulezet.