io.github.feedthrough/feedthrough

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

  pass25% of grade

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

  Scanned 78 text field(s); no instruction-injection patterns found. PASS = no static red flag. A server can serve different tool descriptions at runtime (the rug-pull class), which static scanning cannot see.

• Supply chain

  warn25% of grade

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

  package name "@feedthrough/mcp" is edit-distance 2 from popular package "yup" (typosquat candidate)

• Credential hygiene

  pass15% of grade

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

  No hardcoded secrets, secret-shaped tool parameters, or leaked keys found. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• Permission scope

  pass15% of grade

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

  No shell-exec, unbounded/sensitive filesystem, or download-and-run primitives found in scanned source. PASS = no static red flag. Static scanning is blind to code fetched or executed at runtime.

• 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 0.3.2 (minor bump): new tool "get_instructions" touches a sensitive surface (email/exec/net/fs); new tool "connection_status" touches a sensitive surface (email/exec/net/fs); new tool "get_console_logs" touches a sensitive surface (email/exec/net/fs); new tool "get_network_requests" touches a sensitive surface (email/exec/net/fs); new tool "query_dom" touches a sensitive surface (email/exec/net/fs); new tool "inspect_element" touches a sensitive surface (email/exec/net/fs); new tool "click" touches a sensitive surface (email/exec/net/fs); new tool "fill" touches a sensitive surface (email/exec/net/fs); new tool "hover" touches a sensitive surface (email/exec/net/fs); new tool "press_key" touches a sensitive surface (email/exec/net/fs); new tool "get_html" touches a sensitive surface (email/exec/net/fs); new tool "get_page_info" touches a sensitive surface (email/exec/net/fs); new tool "set_style" appeared; new tool "set_attribute" appeared; new tool "set_text" appeared; new tool "reset_overrides" appeared.

• Transport config

  unscannable10% of grade

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

  local stdio server: no remote transport or auth surface, and no phone-home/telemetry signal found in source. Absence of transport findings is NOT a safety guarantee — phone-home hosts can be constructed at runtime and defeat static analysis.