Indirect Prompt Injection Vulnerability in WeKnora Framework by Tencent
CVE-2026-30856

5.9MEDIUM

Key Information:

Vendor

Tencent

Status
Weknora
Vendor
CVE Published:
7 March 2026

What is CVE-2026-30856?

The WeKnora framework, which leverages LLM for deep document understanding and semantic retrieval, has a vulnerability that stems from tool name collisions and indirect prompt injection. This flaw allows malicious remote MCP servers to hijack legitimate tool executions by exploiting ambiguous naming conventions within the MCP client. Specifically, an attacker can register a malicious tool that overwrites a legitimate one, such as 'tavily_extract', thus redirecting the execution flow and potentially exfiltrating sensitive system prompts and context, while executing other tools with user privileges. This vulnerability has been addressed in the latest update, version 0.3.0.

Human OS v1.0:
Ageing Is an Unpatched Zero-Day Vulnerability.

Remediate biological technical debt. Prime Ageing uses 95% high-purity SIRT6 activation to maintain genomic integrity and bolster systemic resilience.

Deploy the Patch

Affected Version(s)

WeKnora < 0.3.0

References

https://github.com/Tencent/WeKnora/security/advisories/GH...
x_refsource_CONFIRM

CVSS V3.1

Score:
5.9
Severity:
MEDIUM
Confidentiality:
High
Integrity:
Low
Availability:
High
Attack Vector:
Network
Attack Complexity:
High
Privileges Required:
Low
User Interaction:
Required
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

JSON
.