CVE-2026-75130Public exploitNetwork attack vectorHinoki check available

Agent instruction execution via poisoned Custom Rules

Published Aug 18, 2026 · Updated Aug 19, 2026

Prompt injection in Upstash Context7 through 2.1.2 allows attackers to execute instructions in AI coding agents via poisoned Custom AI Instructions. The library-owner Custom Rules field is returned verbatim in MCP query-docs responses, so untrusted rule text is mixed with trusted documentation and treated by the agent as actionable guidance. Exploitation requires control of a submitted or claimed library entry and a victim agent that queries it with file, shell, or network tools enabled; the server itself stays read-only, but the downstream agent can exfiltrate secrets and delete local files.

CVSS severity6.4
Medium
EPSS probability0.28%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkAvailable
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

Prompt injection in Upstash Context7 through 2.1.2 allows attackers to execute instructions in AI coding agents via poisoned Custom AI Instructions. The library-owner Custom Rules field is returned verbatim in MCP query-docs responses, so untrusted rule text is mixed with trusted documentation and treated by the agent as actionable guidance. Exploitation requires control of a submitted or claimed library entry and a victim agent that queries it with file, shell, or network tools enabled; the server itself stays read-only, but the downstream agent can exfiltrate secrets and delete local files.

The record

CVE
CVE-2026-75130
Published
Aug 18, 2026
Updated
Aug 19, 2026
Vendor
Upstash
Product
Context7
Classifications
CWE-1427, T1195.001
Attack vector
network
Privileges
authenticated

Timeline

How it unfolded

  1. Aug 18, 2026CVE publishedPublication date reported by the CVE source.
  2. Aug 19, 2026Record updatedLatest update available in the CVE record.

Exploitability

Present is not the same as exploitable

Compare your product and version with the public record. A matching version still requires validation against your environment.

Is a vulnerable build present?

Compare these published version ranges with your installed build and any vendor patches.

  1. Affected versionversion=0 <=2.1.2

What conditions does exploitation require?

Attack vectornetwork
Required privilegesauthenticated

What is affected?

Upstash · Context7version=0 <=2.1.2

Published CVSS scores

9.0VulnCheckCVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:C/C:H/I:H/A:H
6.4VulnCheckCVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:P/VC:N/VI:N/VA:N/SC:H/SI:H/SA:H

CVSS describes severity. EPSS estimates exploitation probability.

Attacks

What attackers are doing with it

Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 10, 2026Sep 16, 2026
Latest reporting daySep 16, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesauthenticated
Known exploitationUnconfirmed
Public exploitPublished

Weakness, pattern, technique

CWE-1427Improper Neutralization of Input Used for LLM Prompting
T1195.001Compromise Software Dependencies and Development Tools

Public exploit references

Labels summarize the accepted research assessment. They do not indicate a test against your environment.

Technologies

Your stack

See the directory against your own environment.

Your stack

Check the software in your environment

Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.

Book a demo