# Graphiti vulnerable to Cypher Injection via unsanitized node_labels in search filters

Advisory record · `GHSA-gg5m-55jj-8m5g`

- **Identifiers:** `GHSA-gg5m-55jj-8m5g` · `CVE-2026-32247` · `PYSEC-2026-2504`
- **State:** Confirmed
- **Severity:** CVSS 8.1 High
- **Critical:** No
- **Weakness (CWE):** [CWE-943](https://cwe.mitre.org/data/definitions/943.html)
- **Tags:** [Other injection](https://agentprime.io/advisories/tag/injection.md)
- **Published:** 12 Mar 2026
- **Updated:** 13 Jul 2026
- **Reviewed by a person:** No

Published by the GitHub Advisory Database at 12 Mar 2026, 17:26 UTC. agentprime has not analysed this yet, so whether it affects any app is not known.

## Affected packages

- `graphiti-core` (PyPI): affected `<0.28.2`; fixed in `0.28.2`

## Machine-written summary

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

> Graphiti (PyPI graphiti-core) versions before 0.28.2 are vulnerable to Cypher injection through node label filters on non-Kuzu backends such as Neo4j, FalkorDB, and Neptune. An attacker who can supply label values, directly or by prompt-injecting an LLM client that uses the Graphiti MCP server, could run arbitrary Cypher with the database connection's privileges, reading, modifying, or deleting graph data. Upgrade to graphiti-core 0.28.2 or later. Until then, avoid passing untrusted values into node label filters or MCP entity_types, and keep the MCP tools away from untrusted prompts.

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

- 8.1 High, CVSS 3.1: `CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N`, from CNA (GitHub_M)

- **CISA KEV:** Not listed.
- **EPSS:** 0.0042, the 36th percentile, on 14 Nov 2026: FIRST’s estimate of the chance of exploitation in the next 30 days.

## Timeline

- 31 Oct 2026, 15:01 UTC: First seen by agentprime
- 31 Oct 2026, 15:01 UTC: [Confirmed by OSV](https://osv.dev/vulnerability/GHSA-gg5m-55jj-8m5g)

## Sources

- [GitHub Advisory Database: GHSA-gg5m-55jj-8m5g](https://github.com/advisories/GHSA-gg5m-55jj-8m5g)
- [OSV: GHSA-gg5m-55jj-8m5g](https://osv.dev/vulnerability/GHSA-gg5m-55jj-8m5g)
- <https://github.com/getzep/graphiti/security/advisories/GHSA-gg5m-55jj-8m5g>
- <https://nvd.nist.gov/vuln/detail/CVE-2026-32247>
- <https://github.com/getzep/graphiti/pull/1312>
- <https://github.com/getzep/graphiti/commit/7d65d5e77e89a199a62d737634eaa26dbb04d037>
- <https://github.com/getzep/graphiti>
- <https://github.com/getzep/graphiti/releases/tag/v0.28.2>

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-gg5m-55jj-8m5g>, or `POST https://api.agentprime.io/v1/match` with your packages (<https://agentprime.io/api.md>)

---

The web page: <https://agentprime.io/advisories/GHSA-gg5m-55jj-8m5g/>. Everything for agents: <https://agentprime.io/llms.txt>.