Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐍 PyPI

GHSA-c5cp-vx83-jhqx

Langflow Missing Authentication on Critical API Endpoints

Also known asCVE-2026-21445
Published
Jan 2, 2026
Updated
Feb 3, 2026
Affected
2 pkgs
Patched
2 / 2
Exploits
None indexed

EPSS Exploitation Probability

via FIRST.org ↗
21.3%probability of exploitation in next 30 days
Moderate Risk97th percentile+12.25%
0.00%9.21%18.4%27.6%0.1%21.3%Feb 26May 26Jun 26

EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.

Blast Radius

2 pkgs affected
🐍langflow-base🐍langflow

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Summary

Multiple critical API endpoints in Langflow are missing authentication controls, allowing any unauthenticated user to access sensitive user conversation data, transaction histories, and perform destructive operations including message deletion. This affects endpoints handling personal data and system operations that should require proper authorization.

Details

The vulnerability exists in three API endpoints within src/backend/base/langflow/api/v1/monitor.py that are missing the required dependencies=[Depends(get_current_active_user)] authentication dependency:

Affected Endpoints:

  1. GET /api/v1/monitor/messages (Line 61)

    @router.get("/messages")  # ❌ Missing authentication
    async def get_messages(
        session: DbSession,
        flow_id: Annotated[UUID | None, Query()] = None,
        session_id: Annotated[str | None, Query()] = None,
        # ... other parameters
    ) -> list[MessageResponse]:
    
  2. GET /api/v1/monitor/transactions (Line 183)

    @router.get("/transactions")  # ❌ Missing authentication
    async def get_transactions(
        flow_id: Annotated[UUID, Query()],
        session: DbSession,
        params: Annotated[Params | None, Depends(custom_params)],
    ) -> Page[TransactionTable]:
    
  3. DELETE /api/v1/monitor/messages/session/{session_id} (Line 165)

    @router.delete("/messages/session/{session_id}", status_code=204)  # ❌ Missing authentication
    async def delete_messages_session(
        session_id: str,
        session: DbSession,
    ):
    

Inconsistency Evidence: Other endpoints in the same file properly implement authentication:

@router.get("/messages/sessions", dependencies=[Depends(get_current_active_user)])  # ✅ Properly secured
@router.delete("/messages", status_code=204, dependencies=[Depends(get_current_active_user)])  # ✅ Properly secured

PoC

Complete reproduction steps to demonstrate the vulnerability:

Prerequisites:

  1. Start a Langflow server instance
  2. Ensure no authentication headers or API keys are provided

Reproduction Commands:

# 1. Access all user conversations without authentication
curl http://localhost:7860/api/v1/monitor/messages

# 2. Access transaction history without authentication
curl "http://localhost:7860/api/v1/monitor/transactions?flow_id=00000000-0000-0000-0000-000000000000"

# 3. Delete user messages by session without authentication
curl -X DELETE http://localhost:7860/api/v1/monitor/messages/session/00000000-0000-0000-0000-000000000000

Expected vs Actual Behavior:

  • Expected: All requests should return 401 Unauthorized
  • Actual: All requests return successful responses with sensitive data or perform destructive operations

Impact

Vulnerability Type: Broken Authentication and Authorization (OWASP Top 10 - A01:2021)

Severity: High

Who is Impacted:

  • All Langflow users: Personal conversation data exposed to unauthorized access
  • System administrators: Transaction logs disclosed

Specific Impacts:

  1. Data Breach: Unauthorized access to user conversations containing potentially sensitive personal information
  2. Privacy Violation: Transaction histories and user activity patterns exposed without consent
  3. Data Destruction: Malicious actors can delete user conversation histories without authorization
  4. Compliance Risk: Potential violations of data protection regulations (GDPR, CCPA, etc.)
  5. System Intelligence: Attackers can gather information about system usage patterns and user behavior

Attack Scenarios:

  • Malicious users accessing proprietary conversation data
  • Malicious users deleting other users' conversation histories
  • Automated scraping of all user conversations for data harvesting
  • Reconnaissance attacks to understand system architecture and usage patterns

Recommended Fix: Add authentication dependencies to all affected endpoints:

@router.get("/messages", dependencies=[Depends(get_current_active_user)])
@router.get("/transactions", dependencies=[Depends(get_current_active_user)])
@router.delete("/messages/session/{session_id}", dependencies=[Depends(get_current_active_user)])

Environment:

  • Langflow Version: Current main branch
  • Affected Components: API v1 monitoring endpoints
  • Authentication System: FastAPI dependency injection with get_current_active_user
<img width="1908" height="1029" alt="25-090901" src="https://github.com/user-attachments/assets/44bd03b4-6ada-45b7-b81b-9cb83747172b" />

Affected Packages

2 total 2 fixed
EcosystemPackageVulnerable rangeFix
🐍PyPIlangflow-baseall versions0.7.1
🐍PyPIlangflowall versions1.7.1

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for langflow-base. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.

  2. Fix

    Update langflow-base to 0.7.1 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-c5cp-vx83-jhqx is resolved across your whole dependency graph.

  3. Workarounds

    If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.

  4. How O3 protects you

    O3 pinpoints whether GHSA-c5cp-vx83-jhqx is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.

Tailored to GHSA-c5cp-vx83-jhqx. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

### Summary Multiple critical API endpoints in Langflow are missing authentication controls, allowing any unauthenticated user to access sensitive user conversation data, transaction histories, and perform destructive operations including message deletion. This affects endpoints handling personal data and system operations that should require proper authorization. ### Details The vulnerability exists in three API endpoints within `src/backend/base/langflow/api/v1/monitor.py` that are missing the required `dependencies=[Depends(get_current_active_user)]` authentication dependency: **Affected
O3 Security · Impact-Aware SCA

Is GHSA-c5cp-vx83-jhqx in your dependencies?

O3 detects GHSA-c5cp-vx83-jhqx across PyPI dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.