GHSA-2XJ6-XX86-CWWC
Vulnerability from github – Published: 2026-09-22 20:34 – Updated: 2026-09-22 20:34Summary
A critical Confused Deputy (Arbitrary File Read & Exfiltration) vulnerability in the Atlassian MCP server (Python) allows an AI agent to exfiltrate sensitive host files and environment secrets. By providing absolute system paths to the attachments parameter of the update_issue tool, an agent can force the privileged MCP process to read and upload any file it has access to—including its own environment variables and host configuration—directly to a Jira ticket.
Details
The vulnerability is located in the update_issue tool handler. The attachments parameter accepts a list of file paths that are passed directly to the Jira API's upload method without any sanitization or validation.
Specifically, the implementation fails to: 1. Restrict paths to a safe workspace: It accepts absolute paths starting with /. 2. Prevent directory traversal: It does not filter for ../ sequences. 3. Validate ownership: It allows the process to read system-level files (like /proc/self/environ) that the calling agent is normally forbidden from accessing due to sandbox restrictions.
Because the MCP server typically runs with higher privileges than the AI agent (holding the JIRA_API_TOKEN and having wider filesystem access), it acts as a "Confused Deputy," performing exfiltration on behalf of a restricted agent.
PoC
To reproduce the vulnerability in an environment where the agent is sandboxed (e.g., as a restricted Unix user) but the MCP server has host-level access:
1. Exfiltrate MCP Secrets: The agent calls the update_issue tool with the following payload: { "issue_key": "SECURITY-1", "attachments": ["/proc/self/environ"] } Result: The MCP server reads its own process environment—containing the JIRA_API_TOKEN and other sensitive keys in plain text—and attaches it to the Jira issue.
2. Exfiltrate Host Configuration: The agent calls the tool to target global assistant settings: { "issue_key": "SECURITY-1", "attachments": ["/home/hermes/.hermes/config.yaml"] } Result: The host's global configuration file (containing tokens for GitLab, Slack, and other services) is successfully exfiltrated to the cloud.
Impact
* Critical Credential Theft: Immediate exposure of the JIRA_API_TOKEN and any other secrets injected into the MCP server's environment. * Arbitrary File Access: Total exfiltration of any data the host user running the MCP server can read (SSH keys, databases, local source code). * Privilege Escalation & Lateral Movement: Attackers can move from a restricted AI sandbox to full host-system control by stealing tokens or configuration files.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "mcp-atlassian"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "0.22.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-77255"
],
"database_specific": {
"cwe_ids": [
"CWE-22",
"CWE-441"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-22T20:34:52Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "### Summary\nA critical Confused Deputy (Arbitrary File Read \u0026 Exfiltration) vulnerability in the Atlassian MCP server (Python) allows an AI agent to exfiltrate sensitive host files and environment secrets. By providing absolute system paths to the attachments parameter of the update_issue tool, an agent can force the privileged MCP process to read and upload any file it has access to\u2014including its own environment variables and host configuration\u2014directly to a Jira ticket.\n\n### Details\nThe vulnerability is located in the update_issue tool handler. The attachments parameter accepts a list of file paths that are passed directly to the Jira API\u0027s upload method without any sanitization or validation. \n\nSpecifically, the implementation fails to:\n1.\u00a0 Restrict paths to a safe workspace: It accepts absolute paths starting with /.\n2.\u00a0 Prevent directory traversal: It does not filter for ../ sequences.\n3.\u00a0 Validate ownership: It allows the process to read system-level files (like /proc/self/environ) that the calling agent is normally forbidden from accessing due to sandbox restrictions.\n\nBecause the MCP server typically runs with higher privileges than the AI agent (holding the JIRA_API_TOKEN and having wider filesystem access), it acts as a \"Confused Deputy,\" performing exfiltration on behalf of a restricted agent.\n\n### PoC\nTo reproduce the vulnerability in an environment where the agent is sandboxed (e.g., as a restricted Unix user) but the MCP server has host-level access:\n\n1.\u00a0 Exfiltrate MCP Secrets:\n\u00a0\u00a0\u00a0\u00a0The agent calls the update_issue tool with the following payload:\n {\n \"issue_key\": \"SECURITY-1\",\n \"attachments\": [\"/proc/self/environ\"]\n }\n \u00a0\u00a0\u00a0\u00a0Result: The MCP server reads its own process environment\u2014containing the JIRA_API_TOKEN and other sensitive keys in plain text\u2014and attaches it to the Jira issue.\n\n2.\u00a0 Exfiltrate Host Configuration:\n\u00a0\u00a0\u00a0\u00a0The agent calls the tool to target global assistant settings:\n {\n \"issue_key\": \"SECURITY-1\",\n \"attachments\": [\"/home/hermes/.hermes/config.yaml\"]\n }\n \u00a0\u00a0\u00a0\u00a0Result: The host\u0027s global configuration file (containing tokens for GitLab, Slack, and other services) is successfully exfiltrated to the cloud.\n\n### Impact\n*\u00a0\u00a0 Critical Credential Theft: Immediate exposure of the JIRA_API_TOKEN and any other secrets injected into the MCP server\u0027s environment.\n*\u00a0\u00a0 Arbitrary File Access: Total exfiltration of any data the host user running the MCP server can read (SSH keys, databases, local source code).\n*\u00a0\u00a0 Privilege Escalation \u0026 Lateral Movement: Attackers can move from a restricted AI sandbox to full host-system control by stealing tokens or configuration files.",
"id": "GHSA-2xj6-xx86-cwwc",
"modified": "2026-09-22T20:34:52Z",
"published": "2026-09-22T20:34:52Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/sooperset/mcp-atlassian/security/advisories/GHSA-2xj6-xx86-cwwc"
},
{
"type": "WEB",
"url": "https://github.com/sooperset/mcp-atlassian/pull/1448"
},
{
"type": "WEB",
"url": "https://github.com/sooperset/mcp-atlassian/commit/b041733473f95119dd539542a43c280737a8e460"
},
{
"type": "PACKAGE",
"url": "https://github.com/sooperset/mcp-atlassian"
},
{
"type": "WEB",
"url": "https://github.com/sooperset/mcp-atlassian/releases/tag/v0.22.0"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N",
"type": "CVSS_V3"
}
],
"summary": "MCP Atlassian: Arbitrary File Read \u0026 Exfiltration (Confused Deputy) in JIRA update_issue"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.