# Langroid: SQLChatAgent dangerous-function blocklist can be bypassed with quoted or schema-qualified pg_read_file calls

Advisory record · `GHSA-6xc5-4r68-67fc`

- **Identifiers:** `GHSA-6xc5-4r68-67fc` · `CVE-2026-54760` · `PYSEC-2026-2577`
- **State:** Confirmed
- **Severity:** CVSS 9.3 Critical
- **Critical:** Yes: CVSS 9+
- **Weakness (CWE):** [CWE-22](https://cwe.mitre.org/data/definitions/22.html) · [CWE-89](https://cwe.mitre.org/data/definitions/89.html)
- **Tags:** [Path traversal](https://agentprime.io/advisories/tag/path-traversal.md) · [SQL injection](https://agentprime.io/advisories/tag/sql-injection.md)
- **Published:** 6 Jul 2026
- **Updated:** 13 Jul 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 6 Jul 2026, 20:39 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Critical

Critical, because:

- CVSS score 9.3: 9.0 or more.

How we decide what’s critical: <https://agentprime.io/advisories/critical.md>

## Affected packages

- `langroid` (PyPI): affected `<0.65.1`; fixed in `0.65.1`

## Machine-written summary

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

> GHSA-6xc5-4r68-67fc concerns langroid before 0.65.1, where the SQLChatAgent dangerous-function regex can be bypassed using quoted identifiers, inline comments, or schema-qualified names in PostgreSQL calls to pg_read_file. An attacker who can influence the SQL that SQLChatAgent generates can use these forms to read server-side files, provided the database role can call pg_read_file. Upgrade to langroid 0.65.1 or later; a least-privilege database role without pg_read_server_files also reduces exposure.

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

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

- **CISA KEV:** Not listed.
- **EPSS:** 0.0065, the 51st percentile, on 14 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 15:02 UTC: First seen by agentprime
- 31 Oct 2026, 15:02 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-6xc5-4r68-67fc)

## Sources

- [GitHub Advisory Database: GHSA-6xc5-4r68-67fc](https://github.com/advisories/GHSA-6xc5-4r68-67fc)
- [OSV: GHSA-6xc5-4r68-67fc](https://osv.dev/vulnerability/GHSA-6xc5-4r68-67fc)
- <https://github.com/langroid/langroid/security/advisories/GHSA-6xc5-4r68-67fc>
- <https://github.com/langroid/langroid>

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-6xc5-4r68-67fc>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-6xc5-4r68-67fc/>. Everything for agents: <https://agentprime.io/llms.txt>.