AAK-MCP-SEQTHINKING-CVE-2026-81845-001: mcp-sequential-thinking 0.5.0 or earlier (session import/export path traversal)
- Severity
- MEDIUM
- Category
- Supply Chain (
supply-chain) - Scanner
agent_audit_kit/scanners/mcp_cve_pins_2026_07.py- Auto-fixable
- No.
- SARIF name
McpSequentialThinkingSessionPathTraversal
What it detects
mcp-sequential-thinking (PyPI, arben-adm) up to and including 0.5.0 takes a file_path argument on the import_session and export_session tools in mcp_sequential_thinking/server.py and does not confine it to a session directory, so a caller reaches arbitrary paths on the host — reading files back through an import, or writing them through an export. Both are ordinary MCP tools, so the caller is whatever can drive the server, which in an agent pipeline includes anything that can steer tool selection. Fixed in 0.6.0 (patch 2fad3ee); the exploit is public. Treat <= 0.5.0 and unpinned as exposed.
How to fix
Upgrade mcp-sequential-thinking to >= 0.6.0 and pin it. If you cannot upgrade yet, disable the import_session / export_session tools rather than relying on the caller to send a well-behaved path, and run the server as a user with no read access to anything outside its session directory.
To confirm the finding is gone, report only this rule:
agent-audit-kit scan . --rules AAK-MCP-SEQTHINKING-CVE-2026-81845-001
Limitations
Reads mcp-sequential-thinking in requirements*.txt at any depth and, at the project root only, in pyproject.toml, Pipfile, package.json, .mcp.json, mcp.json, claude_desktop_config.json and the Pipfile.lock, poetry.lock, uv.lock, package-lock.json, pnpm-lock.yaml and yarn.lock lockfiles, taking the lowest version a range allows and reporting a reference that states no version. The dependency is not read when it is declared in a sub-directory manifest (services/api/pyproject.toml), in .cursor/mcp.json or .vscode/mcp.json, or in a lockfile format the reader does not parse (pdm.lock). A file over 2 MB, or a lockfile over 32 MiB, is not read. A declaration nested more than 64 levels deep in package.json, a TOML manifest or an MCP config (an MCP server more than 62) is still judged, and reported without a line; a lockfile entry keeps its line at any depth.
References
- CVE: CVE-2026-81845
- OWASP MCP Top 10: MCP04:2025 (Software Supply Chain Attacks & Dependency Tampering), MCP05:2025 (Command Injection & Execution)
- OWASP Agentic Top 10: ASI02 (Tool Misuse and Exploitation), ASI04 (Agentic Supply Chain Vulnerabilities)
- Adversa AI MCP Top 25 (AAK-defined ids): ADV-AUTH-01 (Missing Authentication)
- CSA AI Controls Matrix (AICM): AIS-08, STA-08
All rules. This page is generated from the rule registry when the documentation is built; to change it, change the rule in agent_audit_kit/rules/builtin.py. To give the rule a hand-written write-up, add docs/rules/AAK-MCP-SEQTHINKING-CVE-2026-81845-001.md: the page then shows that text, with the registry reference appended under it.