# PraisonAI: Unsafe Dynamic Module Loading Leads to Arbitrary Code Execution via tools.py in AgentFlow

Advisory record · `GHSA-4gfv-wg42-7jw5`

- **Identifiers:** `GHSA-4gfv-wg42-7jw5` · `CVE-2026-61437` · `PYSEC-2026-4267`
- **State:** Confirmed
- **Severity:** CVSS 8.5 High
- **Critical:** No
- **Weakness (CWE):** [CWE-693](https://cwe.mitre.org/data/definitions/693.html) · [CWE-829](https://cwe.mitre.org/data/definitions/829.html)
- **Tags:** [CWE-693](https://agentprime.io/advisories/tag/cwe-693.md) · [CWE-829](https://agentprime.io/advisories/tag/cwe-829.md)
- **Published:** 8 Oct 2026
- **Updated:** 15 Oct 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 8 Oct 2026, 16:48 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `praisonaiagents` (PyPI): affected `<1.6.78`; fixed in `1.6.78`

## Machine-written summary

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

> The praisonaiagents package before 1.6.78 is vulnerable to arbitrary Python code execution. When a workflow step uses a string output_pydantic reference, the framework automatically imports and runs a tools.py file located next to the workflow file, without the environment-variable protections used elsewhere. An attacker who can place a workflow file and a matching tools.py in a location the application loads from could run code with the privileges of the workflow process. Upgrade to praisonaiagents 1.6.78 or later.

Written 1 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

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

- **CISA KEV:** Not listed.
- **EPSS:** 0.0020, the 9th percentile, on 5 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 14:08 UTC: First seen by agentprime
- 31 Oct 2026, 14:08 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-4gfv-wg42-7jw5)

## Sources

- [GitHub Advisory Database: GHSA-4gfv-wg42-7jw5](https://github.com/advisories/GHSA-4gfv-wg42-7jw5)
- [OSV: GHSA-4gfv-wg42-7jw5](https://osv.dev/vulnerability/GHSA-4gfv-wg42-7jw5)
- <https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4gfv-wg42-7jw5>
- <https://nvd.nist.gov/vuln/detail/CVE-2026-61437>
- <https://github.com/MervinPraison/PraisonAI>
- <https://www.vulncheck.com/advisories/praisonai-before-remote-code-execution-via-tools-py>

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-4gfv-wg42-7jw5>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-4gfv-wg42-7jw5/>. Everything for agents: <https://agentprime.io/llms.txt>.