🚀 The fast, Pythonic way to build MCP servers and clients.
E critical
- vulnerabilities attested + 50
- capability exposure inferred + 35
- recent drift inferred + 12
- tool safety inferred + 12
- 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 9m ago · see ecosystem CVEs →
- D · 77 → E · 100
- E · 89 → D · 77
- E · 97 → E · 89
- E · 100 → E · 97
- E · 94 → E · 100
- C · 42 → E · 94
Inferred surface — each links to servers holding it:
HIGH
Missing Consent Verification in OAuth Proxy Callback Facilitates Confused Deputy Vulnerabilities
EPSS 0%
HIGH
SSRF & Path Traversal Vulnerability in FastMCP OpenAPI Provider
EPSS 1%
HIGH
FastMCP Auth Integration Allows for Confused Deputy Account Takeover
MEDIUM
Command injection via server name in subprocess-backed install commands
EPSS 1%
MEDIUM
Improper Resource Handling in FastMCP OAuth Proxy Enables Token Reuse Across MCP Servers
EPSS 0%
LOW
Update to MCP 1.23+
MEDIUM
Windows command injection in FastMCP Cursor installer via server_name
EPSS 0%
MEDIUM
Reflected XSS in client's callback page
EPSS 0%
- high dangerous code
dynamic exec: __import__(), eval()/exec(), unsafe yaml.load()
prompt-surface
shipped agent-instruction files + hidden-content / dangerous-code findings —
quoted from the analyzed source
analyzed commit 44c0907 · analyzer v28 · 2d ago
skills & prompt files 13
- skill PrefectHQ-fastmcp-44c0907/.claude/skills/code-review/SKILL.md
- skill PrefectHQ-fastmcp-44c0907/.claude/skills/python-tests/SKILL.md
- skill PrefectHQ-fastmcp-44c0907/.claude/skills/review-issue/SKILL.md
- skill PrefectHQ-fastmcp-44c0907/.claude/skills/review-pr/SKILL.md
- agent-rules PrefectHQ-fastmcp-44c0907/.cursor/rules/core-mcp-objects.mdc
- agent-rules PrefectHQ-fastmcp-44c0907/CLAUDE.md
- agent-rules PrefectHQ-fastmcp-44c0907/docs/.cursor/rules/mintlify.mdc
- agent-rules PrefectHQ-fastmcp-44c0907/docs/integrations/gemini.mdx
- agent-rules PrefectHQ-fastmcp-44c0907/docs/v2/integrations/gemini.mdx
- agent-rules PrefectHQ-fastmcp-44c0907/docs/v3/integrations/gemini.mdx
- skill PrefectHQ-fastmcp-44c0907/examples/skills/sample_skills/code-review/SKILL.md
- skill PrefectHQ-fastmcp-44c0907/examples/skills/sample_skills/pdf-processing/SKILL.md
- skill PrefectHQ-fastmcp-44c0907/skills/fastmcp-client-cli/SKILL.md
danger signals7
- dynamic code execution __import__() PrefectHQ-fastmcp-44c0907/tests/docs/test_doc_examples.py :79
__import__(alias.name) - dynamic code execution eval()/exec() PrefectHQ-fastmcp-44c0907/tests/docs/test_upgrade_guide_equivalence.py :45
exec(compile(matches[0].source, str(path), "exec"), namespace) - dynamic code execution eval()/exec() PrefectHQ-fastmcp-44c0907/tests/docs/test_upgrade_guide_examples.py :101
exec(compile(example.source, str(example.path), "exec"), namespace) - dynamic code execution eval()/exec() PrefectHQ-fastmcp-44c0907/tests/experimental/transforms/test_code_mode.py :112
exec(wrapped, namespace, namespace) - dynamic code execution eval()/exec() PrefectHQ-fastmcp-44c0907/tests/experimental/transforms/test_code_mode_discovery.py :76
exec(wrapped, namespace, namespace) - dynamic code execution unsafe yaml.load() PrefectHQ-fastmcp-44c0907/tests/utilities/json_schema_type/test_real_world_schemas.py :204
spec = yaml.load(spec_file.read_text(), Loader=CSafeLoader) - suspicious endpoint 169.254.169.254 (cloud metadata) PrefectHQ-fastmcp-44c0907/tests/utilities/test_json_schema.py :1062
"evil": {"$ref": "http://169.254.169.254/latest/meta-data/"},
- recent drift +12 capability drift →
[](https://mcpobservatory.com/servers/github:PrefectHQ/fastmcp/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 PrefectHQ.