# Cortex has Untrusted Project Bootstrap Code Execution via \`CLAUDE_PROJECT_DIR\`

Advisory record · `GHSA-gvpp-v77h-5w8g`

- **Identifiers:** `GHSA-gvpp-v77h-5w8g` · `CVE-2026-49986` · `PYSEC-2026-2676`
- **State:** Confirmed
- **Severity:** CVSS 7.1 High
- **Critical:** No
- **Weakness (CWE):** [CWE-829](https://cwe.mitre.org/data/definitions/829.html)
- **Tags:** [CWE-829](https://agentprime.io/advisories/tag/cwe-829.md)
- **Published:** 1 Jul 2026
- **Updated:** 13 Jul 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 1 Jul 2026, 18:41 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `neuro-cortex-memory` (PyPI): affected `<3.18.0`; fixed in `3.18.0`

## Machine-written summary

Summary written by Claude Haiku from the source advisory. Not reviewed by a person.

> Versions of neuro-cortex-memory before 3.18.0 are affected. The Cortex visualization tool treats the directory a user has open in Claude Code as a trusted Cortex source root when it finds two marker files there, so a malicious project can get attacker-chosen Python code run with the victim's local user privileges when the victim invokes the open_visualization tool. Upgrade neuro-cortex-memory to version 3.18.0 or later.

Written 6 Nov 2026.

## Not analysed yet

We haven’t published our analysis of this Advisory yet. Once a person has reviewed it, this page will say what’s vulnerable, what has to be true for your app to be affected, and how to check.

## Severity and exploitation

- 7.1 High, CVSS 4.0: `CVSS:4.0/AV:L/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:P`, from CNA (GitHub_M)

- **CISA KEV:** Not listed.
- **EPSS:** 0.0016, the 5th percentile, on 7 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 15:00 UTC: First seen by agentprime
- 31 Oct 2026, 15:00 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-gvpp-v77h-5w8g)

## Sources

- [GitHub Advisory Database: GHSA-gvpp-v77h-5w8g](https://github.com/advisories/GHSA-gvpp-v77h-5w8g)
- [OSV: GHSA-gvpp-v77h-5w8g](https://osv.dev/vulnerability/GHSA-gvpp-v77h-5w8g)
- <https://github.com/cdeust/Cortex/security/advisories/GHSA-gvpp-v77h-5w8g>
- <https://github.com/cdeust/Cortex>
- <https://github.com/cdeust/Cortex/releases/tag/v3.17.1>

Facts on this page come from the GitHub Advisory Database (CC-BY 4.0), OSV and the other sources listed, as published. We don’t reproduce full advisory descriptions, which can contain exploit code.

## Check whether this affects you

A Match means an installed version is in the affected range above. Most don’t affect the app; agentprime hasn’t analysed this one yet.

- With the agentprime MCP server (<https://api.agentprime.io/mcp>), ask your agent to check the project: `check_lockfile` sends a lockfile, `check_packages` a list of installed packages, and both return the Matches and the upgrades that clear them. `get_advisory` returns this record. Setup: <https://agentprime.io/mcp.md>
- In a browser: <https://agentprime.io/check/> reads your lockfile in the browser, never uploading it.
- Whether it touches your code, with file and line: the Claude Code plugin, <https://agentprime.io/plugin.md>
- As JSON: <https://api.agentprime.io/v1/advisories/GHSA-gvpp-v77h-5w8g>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-gvpp-v77h-5w8g/>. Everything for agents: <https://agentprime.io/llms.txt>.