UTC
Medium

ACP resource_link metadata prompt interpolation allowed prompt-injection content via title/URI fields

CVE-2026-27165 · GHSA-74xj-763f-264w · published 2026-02-21

Severity
Medium
CVSS score
Not scored
Chance of exploitation (EPSS)
Not scored
Fixed in
2026.2.21
What it is

Attackers who can influence ACP resource_link metadata can inject attacker-controlled prompt text into the LLM context, potentially altering model behavior.

What to do

Upgrade openclaw to 2026.2.21 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.

Escape control and delimiter characters in ACP resource_link metadata before prompt interpolation. 6aa11f30924ee08d0e6f36ede4877e44c553b984

Affected
PackageRegistryAffectedFixed in
openclawnpm<=2026.2.19-22026.2.21

The impact, patch and workaround text comes from the advisory, shortened (published by OpenClaw). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].