GHSA-xhcr-cqfr-m3hv: atomic-agents-stack: HTTP MCP catalog accepts cleartext http and spawns catalog-supplied commands (MITM to RCE)
- Identifier
- GHSA-xhcr-cqfr-m3hv
- Published
- Record updated
- Affected
- atomic-agents-stack <= 1.0.0
- Fixed in
- 1.1.0
- released June 10, 2026, 68 days before disclosure
Summary
The HTTP MCP server-registry backend, `atomic_agents/mcp_registry/http.py` (`make_http_mcp_server_registry_backend_from_url`), accepts both `http` and `https` catalog URLs, affecting all versions through 1.0.0. Catalog `command`/`args` values are type-validated but not content-restricted, and `MCPClientPool` spawns them as local stdio subprocesses. Over a cleartext `http://` catalog, a network man-in-the-middle can rewrite the response to inject arbitrary commands and gain code execution on the agent host without LLM involvement. The `https` path is stated to be sound, and `mcp_allow_fn` defaults to None, so no allowlist applies unless an operator configures one.
Mitigation
Require `https` by default and gate `http://` behind a loud explicit opt-in. Defense-in-depth: allowlist the resolved command basename (or require confirmation) before any registry-sourced subprocess spawn. Document the consequence in spec/36.
Topics
Related items
- MediumCVE-2026-124248: GPTranslate WordPress plugin stored cross-site scripting via comment parameterSimilar attack · NVD/CVE Database
- MediumAmazon: China-Linked Hackers’ AI Agents Exploited Medical Software Zero-DaySimilar attack · SecurityWeek
- MediumAttackers' AI Agents Found a Zero-Day in Medical-Imaging Software and Gained Root AccessSimilar attack · The Hacker News
- HighCVE-2026-134172: audio-separator insecure deserialization of malicious Demucs model filesSimilar attack · NVD/CVE Database
- MediumCVE-2026-133836: hankcs HanLP deserialization via model_file in RNNLanguageModelSimilar attack · NVD/CVE Database