io.github.pzfreo/build123d-mcp

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

• Credential hygiene

  pass15% of grade

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

  Reads secrets from the environment (e.g. src/build123d_mcp/cli.py) — the recommended intake shape. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• 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/build123d_mcp/tools/_bounded.py: subprocess.* call; examples/live_viewer_client.py: outbound HTTP/socket call). has both shell/exec and destructive filesystem writes (src/build123d_mcp/tools/_bounded.py: subprocess.* call; examples/live_viewer_client.py: open(…, 'w'/'a'/'x') write-mode file access). discloses shell/exec capability: 6 files (src/build123d_mcp/tools/_bounded.py, src/build123d_mcp/tools/design_audit.py, src/build123d_mcp/tools/locate.py, +3 more) discloses dynamic code execution: src/build123d_mcp/session.py: uses eval()/exec() discloses dynamic code execution: src/build123d_mcp/tools/library.py: uses eval()/exec() discloses dynamic code execution: src/build123d_mcp/tools/resolve.py: uses eval()/exec() discloses broad/destructive filesystem access: 24 files (examples/live_viewer_client.py, examples/live_viewer_client.py, src/build123d_mcp/_design_audit_subprocess.py, +21 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 0.3.83 (patch bump): new capability primitive(s) in source: child_process, dynamic-eval, fs-write, net, subprocess; new hardcoded outbound host/BCC destination(s): github.com, schemas.microsoft.com, schemas.openxmlformats.org, www.w3.org.

• 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