io.github.james-robson/push-realm

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

  warn25% of grade

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

  server instructions: imperative telling the model it must read/send/attach something.

• Supply chain

  pass25% of grade

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

  repository linkage verified

• Credential hygiene

  fail15% of grade

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

  tool "withdraw_open_issue" has a secret-shaped input parameter "withdraw_token" — secrets must never be tool parameters (the model would supply/handle the credential).

• Permission scope

  unscannable15% of grade

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

  No source files, package.json, or pyproject were fetched; cannot infer capability scope.

• Version behavior

  warn10% of grade

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

  vs prior 4.7.1 (minor bump): tool "search_solutions" description changed (wording only, no injection pattern); tool "submit_open_issue" description changed (wording only, no injection pattern); new tool "withdraw_open_issue" touches a sensitive surface (email/exec/net/fs).

• Transport config

  fail10% of grade

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

  remote MCP endpoint declares NO authentication (unauthenticated remote — nginx-ui/CVE-2026-33032 class)