github Python analyzed 31c1d77

sooperset/mcp-atlassian

github

MCP server for Atlassian tools (Confluence, Jira)

maintainer
sooperset
license
MIT
first seen
2026-05-22
last seen
2026-08-01
releases · 30d
3
short id
risk 99/100 · heuristic grade
E critical
  • vulnerabilities attested + 50
  • capability exposure inferred + 28
  • recent drift inferred + 12
  • tool safety inferred + 17
  • trust mitigators mixed − 8

attested inferred mixed

The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.

graded 8m ago · see ecosystem CVEs →

risk trajectory 5 movements
  • D · 70 E · 99
  • D · 73 D · 70
  • B · 23 D · 73
  • B · 20 B · 23
  • A · 0 B · 20
capability exposure grade factor +28
Inferred surface — each links to servers holding it:
vulnerabilities 20 CVEs · grade factor +50
CRITICAL
Unauthenticated HTTP MCP requests can use globally configured Jira and Confluence credentials affects ["< 0.22.0"]
CRITICAL
[mcp-atlassian] Authentication bypass in HTTP transport: AtlassianOpaqueTokenVerifier accepts any no … affects ["< 0.22.0"]
CRITICAL
Arbitrary file write leading to arbitrary code execution via unconstrained download_path in confluen … EPSS 2% CVE-2026-27825 affects ["<0.17.0"]
HIGH
SSRF Protection Bypass affects ["< 0.22.0"]
HIGH
Incomplete path traversal fix allows intra-CWD module overwrite and RCE (bypass of GHSA-xjgw-4wvw-rg … affects ["< 0.22.0"]
HIGH
Path Traversal / Arbitrary File Read in confluence_upload_attachment MCP tool (incomplete fix of GHS … affects ["< 0.22.0"]
HIGH
SSRF redirect protection missing for basic-auth and OAuth authentication branches affects ["< 0.22.0"]
HIGH
Arbitrary local file READ via unconstrained file_path in upload_attachment (Confluence + Jira) — mir … affects ["< 0.22.0"]
HIGH
Arbitrary file read/exfiltration via upload_attachment missing validate_safe_path() affects ["< 0.22.0"]
HIGH
Arbitrary file read via confluence_upload_attachment allows exfiltration of server credentials affects ["< 0.22.0"]
HIGH
HTTP upload tools accept arbitrary server-local file paths affects ["< 0.22.0"]
HIGH
Arbitrary File Read & Exfiltration (Confused Deputy) in JIRA update_issue affects ["< 0.22.0"]
HIGH
Jira and Confluence attachment upload tools can read arbitrary server-local files affects ["< 0.22.0"]
HIGH
Multiple vulnerabilities: SSRF via icon_url + attachment URL, JQL injection, Stored XSS affects ["< 0.22.0"]
HIGH
MCP HTTP Client Server-Local File Exfiltration via Unvalidated Attachment Upload Path affects ["< 0.22.0"]
HIGH
Arbitrary server-local file upload to Jira/Confluence attachments via unrestricted file_path paramet … affects ["< 0.22.0"]
HIGH
Unauthenticated arbitrary local file read via upload_attachment file_path, chained with missing auth … affects ["< 0.22.0"]
HIGH
ENABLED_TOOLS / Toolset authorization bypass in mcp-atlassian affects ["< 0.22.0"]
HIGH
Incomplete fix for CVE-2026-27826: DNS rebinding bypasses SSRF validation (validated IP not pinned) affects ["< 0.22.0"]
HIGH
DNS-rebinding TOCTOU bypass of the SSRF fix (GHSA-7r34): unauthenticated SSRF to cloud metadata on p … affects ["< 0.22.0"]
tool safety 2 findings · grade factor +17
  1. high dangerous code

    dynamic exec: pickle.loads(), __import__ sink

  2. medium toxic flow (lethal trifecta)

    lethal trifecta reachable across this server's tool + source surface: private-data access + untrusted-content ingestion + network exfil (a leg is proven only in the analyzed source)

skills & danger signals github-tarball
prompt-surface shipped agent-instruction files + hidden-content / dangerous-code findings — quoted from the analyzed source

analyzed commit 31c1d77 · analyzer v28 · 2d ago

skills & prompt files 3

danger signals7

other grade factors evidence elsewhere
embed badge readme-ready
live risk-grade badge preview [![MCP Observatory risk grade](https://mcpobservatory.com/servers/mcpso:sooperset/mcp-atlassian/badge.svg)](https://mcpobservatory.com/servers/mcpso:sooperset/mcp-atlassian/security)

Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of sooperset.