io.github.NeoResearchAI/MCPServer

What we checked

Six static checks, weighted by risk. Every result reflects only what could be observed in the published package and repository — never intent.

• Injection surface

  unscannable25% of grade

  Tool descriptions/manifest scanned for instruction-injection patterns (imperatives at the model, hidden text, 'ignore previous', data-exfil URLs).

  No tool descriptions, server instructions, prompts, or resources were fetched; nothing to scan for injection.

• Supply chain

  fail25% of grade

  Package provenance: namespace verification, repo linkage, maintainer count, account age, postinstall scripts, typosquat distance.

  Python build script executes at build time: subprocess in build script, network call in build script

• Credential hygiene

  warn15% of grade

  How the server takes secrets (env vs plaintext config vs hardcoded); secrets appearing in tool schemas.

  tests/test_system.py: contains a hardcoded OpenAI API key (sk-a…2a (len 27)) — in an example/sample file; verify it is not a real key. Reads secrets from the environment (e.g. src/neo_mcp/action_handlers.py) — the recommended intake shape.

• Permission scope

  warn15% of grade

  Declared tools vs. breadth (filesystem, network, exec); flags shell-exec and unbounded filesystem access.

  has both shell/exec and network egress — capable of download-and-run, though no such path was observed (src/neo_mcp/integrations/providers/github.py: subprocess.* call; src/neo_mcp/backend_client.py: outbound HTTP/socket call). has both shell/exec and destructive filesystem writes (src/neo_mcp/integrations/providers/github.py: subprocess.* call; src/neo_mcp/backend_poller.py: open(…, 'w'/'a'/'x') write-mode file access). discloses shell/exec capability: 5 files (src/neo_mcp/integrations/providers/github.py, src/neo_mcp/login.py, src/neo_mcp/server.py, +2 more) discloses broad/destructive filesystem access: 6 files (src/neo_mcp/backend_poller.py, src/neo_mcp/integrations/_fsutil.py, src/neo_mcp/job_manager.py, +3 more) Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• Version behavior

  fail10% of grade

  Diff of tool definitions between versions; new permissions or changed descriptions in a patch release (the postmark-mcp class).

  vs prior 0.5.9 (patch bump): new capability primitive(s) in source: child_process, fs-write, http-client, net, subprocess; new hardcoded outbound host/BCC destination(s): alpha.heyneo.com, api.anthropic.com, api.github.com, api.openai.com, api.wandb.ai, app.heyneo.so, backend.test, custom.example.com, docs.heyneo.com, example.com, github.com, heyneo.com, heyneo.so, huggingface.co, master.heyneo.com, mcpserver.heyneo.com, openrouter.ai, pypi.org, s3.amazonaws.com, test, testserver, user, www.kaggle.com, x.

• Transport config

  warn10% of grade

  Remote servers: TLS and auth mode (none/token/OAuth). Local servers: whether the manifest indicates it phones home.

  manifest/source references tracking behavior — disclosure