# CWE-1427.

Every Advisory and CVE whose sources give the weakness CWE-1427, newest first.

## Records, newest first

- [OpenHuman: Indirect prompt injection via background email sync executes Composio actions without user approval](https://agentprime.io/advisories/CVE-2026-54142.md)
  `CVE-2026-54142` · CVE record: tinyhumansai openhuman · CVSS 8.7 High · Not analysed yet · Published 5 Nov 2026 · Tags: CWE-1427
- [OpenHuman: MCP tool description injection: untrusted server descriptions execute in agent LLM context without sanitization](https://agentprime.io/advisories/CVE-2026-54140.md)
  `CVE-2026-54140` · CVE record: tinyhumansai openhuman · CVSS 8.3 High · Not analysed yet · Published 5 Nov 2026 · Tags: CWE-1427
- [Secret masking bypass in the environment tool in Strands Agents Tools](https://agentprime.io/advisories/CVE-2026-115415.md)
  `CVE-2026-115415` · CVE record: amazon strands-agents-tools · CVSS 5.9 Medium · Not analysed yet · Published 28 Oct 2026 · Tags: CWE-1427
- [LangAlpha through 2026.09.07 Indirect Prompt Injection RCE via Bash and ExecuteCode](https://agentprime.io/advisories/CVE-2026-109440.md)
  `CVE-2026-109440` · CVE record: ginlix-ai LangAlpha · CVSS 8.7 High · Not analysed yet · Published 20 Oct 2026 · Tags: CWE-1427
- [MoviePilot 3.1.5 Command Injection via ExecuteCommandTool](https://agentprime.io/advisories/CVE-2026-77931.md)
  `CVE-2026-77931` · CVE record: jxxghp MoviePilot · CVSS 8.7 High · Not analysed yet · Published 15 Oct 2026 · Tags: Command injection, CWE-1427
- [Improper Neutralization of Input Used for LLM Prompting in GitLab](https://agentprime.io/advisories/CVE-2026-96893.md)
  `CVE-2026-96893` · CVE record: GitLab GitLab · CVSS 7.3 High · Not analysed yet · Published 13 Oct 2026 · Tags: CWE-1427
- [Vibe-Trading LLM-callable tools permit command execution, code injection, and SSRF](https://agentprime.io/advisories/GHSA-jqmf-mx4f-hfr6.md)
  `GHSA-jqmf-mx4f-hfr6`, `CVE-2026-114107` · PyPI `vibe-trading-ai`: affected `>=0.1.0, <0.1.7`, fixed in `0.1.7` · CVSS 8.7 High · Not analysed yet · Published 2 Oct 2026 · Tags: Command injection, Code injection, Server-side request forgery (SSRF), CWE-116
- [Microsoft Edge for iOS Spoofing Vulnerability](https://agentprime.io/advisories/CVE-2026-70331.md)
  `CVE-2026-70331` · CVE record: Microsoft Microsoft Edge (Chromium-based) and 1 more · CVSS 5.4 Medium · Not analysed yet · Published 28 Aug 2026 · Tags: CWE-1427
- [Consent bypass in python_repl tool via batch kwargs forwarding in Amazon Strands Agents Tools](https://agentprime.io/advisories/CVE-2026-78379.md)
  `CVE-2026-78379` · CVE record: Amazon strands-agents-tools · CVSS 9.2 Critical · Critical (CVSS 9+) · Not analysed yet · Published 25 Aug 2026 · Tags: CWE-1427
- [Context7 2.1.2 Prompt Injection via Custom AI Instructions](https://agentprime.io/advisories/CVE-2026-75130.md)
  `CVE-2026-75130` · CVE record: Uptash Context7 · CVSS 6.4 Medium · Not analysed yet · Published 18 Aug 2026 · Tags: CWE-1427
- [HCL AION is affected by multiple security vulnerabilities.](https://agentprime.io/advisories/CVE-2026-21832.md)
  `CVE-2026-21832` · CVE record: HCL Software AION · CVSS 4.3 Medium · Not analysed yet · Published 13 Aug 2026 · Tags: CWE-1427
- [Prompt injection bypasses shell tool consent gate in Strands Agents Tools](https://agentprime.io/advisories/CVE-2026-18733.md)
  `CVE-2026-18733` · CVE record: AWS strands-agents-tools · CVSS 7.5 High · Not analysed yet · Published 3 Aug 2026 · Tags: CWE-1427
- [Improper Neutralization of Input Used for LLM Prompting in GitLab](https://agentprime.io/advisories/CVE-2026-15077.md)
  `CVE-2026-15077` · CVE record: GitLab GitLab · CVSS 4.3 Medium · Not analysed yet · Published 29 Jul 2026 · Tags: CWE-1427
- [\[Eclipse Theia\] Indirect Prompt Injection via Auto-Loaded Workspace Prompt Template Files in AI Chat](https://agentprime.io/advisories/GHSA-m973-pr9r-hp2w.md)
  `GHSA-m973-pr9r-hp2w`, `CVE-2026-46580` · npm `@theia/ai-chat` and 5 more: affected `<1.71.0`, fixed in `1.71.0` · CVSS 8.4 High · Not analysed yet · Published 18 Jun 2026 · Tags: CWE-829, CWE-1427
- [\[Eclipse Theia\] Indirect Prompt Injection via Adversarial Workspace File and Directory Names in AI Chat](https://agentprime.io/advisories/GHSA-3jww-hxqj-wfq2.md)
  `GHSA-3jww-hxqj-wfq2`, `CVE-2026-44688` · npm `@theia/ai-chat` and 6 more: affected `<1.71.0`, fixed in `1.71.0` · CVSS 8.4 High · Not analysed yet · Published 18 Jun 2026 · Tags: CWE-829, CWE-1427
- [MCP Calculate Server: Prompt Injection to RCE](https://agentprime.io/advisories/CVE-2026-44717.md)
  `CVE-2026-44717` · CVE record: 611711Dark mcp_calculate_server · CVSS 9.8 Critical · Critical (CVSS 9+) · Not analysed yet · Published 15 May 2026 · Tags: Code injection, CWE-1427
- [nnU-Net: Agentic workflow injection in \`.github/workflows/issue-triage.yml\` of \`MIC-DKFZ/nnUNet\`](https://agentprime.io/advisories/CVE-2026-44246.md)
  `CVE-2026-44246` · CVE record: MIC-DKFZ nnUNet · CVSS 7.2 High · Not analysed yet · Published 12 May 2026 · Tags: CWE-1427
- [Multiple vulnerabilities in 1millionbot Millie chatbot](https://agentprime.io/advisories/CVE-2026-4399.md)
  `CVE-2026-4399` · CVE record: 1millionbot Millie chat · CVSS 8.7 High · Not analysed yet · Published 31 Mar 2026 · Tags: CWE-1427
- [Salesforce Agentforce Vibes Extension: Improper Neutralization of Input Used for LLM Prompting vulnerability in Salesforce…](https://agentprime.io/advisories/CVE-2025-64321.md)
  `CVE-2025-64321` · CVE record: Salesforce Agentforce Vibes Extension · CVSS 5.3 Medium · Not analysed yet · Published 4 Nov 2025 · Tags: CWE-1427
- [Salesforce Agentforce Vibes Extension: Improper Neutralization of Input Used for LLM Prompting vulnerability in Salesforce…](https://agentprime.io/advisories/CVE-2025-64320.md)
  `CVE-2025-64320` · CVE record: Salesforce Agentforce Vibes Extension · CVSS 6.5 Medium · Not analysed yet · Published 4 Nov 2025 · Tags: CWE-1427
- [Salesforce Mulesoft Anypoint Code Builder: Improper Neutralization of Input Used for LLM Prompting vulnerability in Salesforce…](https://agentprime.io/advisories/CVE-2025-64318.md)
  `CVE-2025-64318` · CVE record: Salesforce Mulesoft Anypoint Code Builder · CVSS 5.3 Medium · Not analysed yet · Published 4 Nov 2025 · Tags: CWE-1427
- [Salesforce Mulesoft Anypoint Code Builder: Improper Neutralization of Input Used for LLM Prompting vulnerability in Salesforce…](https://agentprime.io/advisories/CVE-2025-10875.md)
  `CVE-2025-10875` · CVE record: Salesforce Mulesoft Anypoint Code Builder · CVSS 6.5 Medium · Not analysed yet · Published 4 Nov 2025 · Tags: CWE-1427
- [Windsurf Prompt Injection via Filename](https://agentprime.io/advisories/CVE-2025-36730.md)
  `CVE-2025-36730` · CVE record: Windsurf Windsurf · CVSS 4.6 Medium · Not analysed yet · Published 14 Oct 2025 · Tags: CWE-1427
- [Improper Neutralization of Input Used for LLM Prompting in GitLab](https://agentprime.io/advisories/CVE-2024-3303.md)
  `CVE-2024-3303` · CVE record: GitLab GitLab · CVSS 6.4 Medium · Not analysed yet · Published 13 Feb 2025 · Tags: CWE-1427

## Filters and formats

Filters are query parameters on this URL: `q` (words, as in a web search engine: `"a phrase"`, `OR`, `-word`; an id; a package or product name), `severity` (low, medium, high or critical), `critical=1`, `kev=1`, `ecosystem` (npm, pypi, packagist, go, maven, rubygems, crates, nuget, hex), `kind=package`, `assigner` (a CNA’s short name), `min_score` and `state`. Each record’s title links to its Markdown page.

- The same list as JSON: <https://api.agentprime.io/v1/advisories?tag=cwe-1427>
- Every list: <https://agentprime.io/advisories.md>, critical: <https://agentprime.io/advisories/critical.md>; by weakness: <https://agentprime.io/advisories/tag/sql-injection.md> and the other tags in <https://agentprime.io/llms.txt>; by technology: `/advisories/technology/<slug>.md` (`fortinet-fortimail`)

---

The web page: <https://agentprime.io/advisories/tag/cwe-1427/>. Everything for agents: <https://agentprime.io/llms.txt>.